University of Southern California Dissertations and Theses

Morphological and microstructural models of typical development in the hippocampus and white matter

(USC Thesis Other)

Morphological and microstructural models of typical development in the hippocampus and white matter

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

MORPHOLOGICAL AND MICROSTRUCTURAL MODELS
OF TYPICAL DEVELOPMENT
IN THE HIPPOCAMPUS AND WHITE MATTER

by
KIRSTEN MARY LYNCH

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA

In Partial Fulfillment of the Requirements for the Degree
DOCTOR OF PHILOSOPHY
(NEUROSCIENCE)

September 2019
Degree Conferral Date: December, 2019
ii

DEDICATION

This dissertation is dedicated to my guardian angel, David Paesani. Thank you for always
believing in me.

iii

ACKNOWLEDGEMENTS

First and foremost, I would like to thank my mentor, Dr. Kristi A. Clark, for all the support,
guidance, opportunities, and camaraderie she has provided throughout my graduate training. Her
curiosity and passion for science has been a source of inspiration over the years and I will forever
be grateful for the opportunity to develop as a scientist under her advisement.
I would also like to thank the members of my dissertation committee, Drs. Arthur Toga
(Committee Chair), Pat Levitt and Yonggang Shi for their advisement throughout graduate school.
Additionally, I would like to thank Dr. Bosco Tjan for his mentorship and encouragement.
Thank you to all the current and former Clark laboratory members for their support and
encouragement along the way. In particular, thanks to Dr. Farshid Sepehrband, Dr. Ryan Cabeen,
Dr. Anthony Krafnick, Dr. Imola MacPhee, Max Orozco, Clio Gonzalez-Zacarias, Hadley
McGregor, Surafael Yared, and Stephen Gonzalez.
Thank you to Dr. Elizabeth Sowell for exposing me to the fascinating world of
developmental neuroimaging, and to current and former Developmental Cognitive Neuroimaging
Laboratory members Drs. Megan Herting, Kristina Uban, and Prapti Gautam for inspiring me to
pursue my PhD and being terrific role models.
Thank you to the USC Neuroscience Graduate Program and in particular, the Program
Manager, Deanna Solórzano, and Director of Student Services, Dawn Burke, for all their help with
logistical support over the years. Thank you to my PhD cohort, Dr. Talia Nir, Louise Menendez,
Dr. Melanie Sweeney, Dr. Dan Rinker, Dr. Rorry Brenner, Katie Zyuzkin and Dr. Brenton Keller
– I could not have found a more supportive graduate family.
On a personal note, I would like to thank my entire family – the Lynchs and Schuberts –
for their endless love and support. In particular, I would like to thank my aunt, Dr. Kristie Reilly,
for exposing me to the beautiful world of neuroscience and planting that seed at such a young age.
Last, but certainly not least, thank you to David Paesani who always encouraged me to follow my
dreams – I could not have done any of this without you.

iv

TABLE OF CONTENTS
Page
Dedication ………………………………………………………………………………...ii
Acknowledgments ……………………………………………………………………….iii
Table of Contents ………………………………………………………………………...iv
List of Publications ……………………………………………………………………….vi
List of Tables …………………………………………………………………………….vii
List of Figures …………………………………………………………………………..viii

Chapter 1: Introduction and Background………………………………………………….9
1.1 Structural brain development………………………………………………….9
1.2 White matter cellular development…………………………………………..11
1.2.1 Myelination………………………………………………………...11
1.2.2 Axon diameter………………………………………………...…...13
1.2.3 Axon number………………………………………………………14
1.3 Hippocampal cellular development………………………………………….16
1.3.1 Hippocampal structure……………………………………………..16
1.3.2 Hippocampal subfield maturation……………………..…………...19
1.4 Structural neuroimaging methods for white matter…………………………21
1.4.1 Diffusion tensor imaging (DTI)……………………………………21
1.4.2 Neurite orientation and dispersion imaging (NODDI)…………….22
1.5 Structural neuroimaging methods for the hippocampus……………………..24
1.5.1 Volumetric approaches…………………………………………….24
1.5.2 Shape analysis……………………………………………………...25
1.6 Summary and overview of chapters………………………………………….25
1.7 References……………………………………………………………………26
Chapter 2: Magnitude and timing of major white matter tract maturation from infancy
through adolescence with NODDI……………………………………………………….40
2.1 Abstract………………………………………………………………………40
2.2 Introduction…………………………………………………………………..41
2.3 Methods………………………………………………………………………43
v

2.4 Results………………………………………………………………………..48
2.5 Discussion……………………………………………………………………53
2.6 References……………………………………………………………………57
2.7 Supplementary material……………………………………………………...64
Chapter 3: Hippocampal shape maturation in childhood and adolescence………………67
3.1 Abstract………………………………………………………………………67
3.2 Introduction…………………………………………………………………..67
3.3 Methods………………………………………………………………………70
3.4 Results………………………………………………………………………..78
3.5 Discussion……………………………………………………………………84
3.6 References……………………………………………………………………91
3.7 Supplementary material…………………………………………………….105
Chapter 4: The effect of BMI on hippocampal shape across childhood………………..109
4.1 Abstract……………………..………………………………………………109
4.2 Introduction…………………………………………………………………109
4.3 Methods…………………………………………………………………..…111
4.4 Results………………………………………………………………………117
4.5 Discussion………………………………………………………………..…117
4.6 References…………………………………………………………………..121
4.7 Supplementary material…………………………………………………….130
Chapter 5: Microstructure-mesh projection: An approach for the analysis of regional
hippocampal microstructure…………………………………………………………….131
5.1 Abstract……………………..………………………………………………131
5.2 Introduction…………………………………………………………………131
5.3 Methods…………………………………………………………………..…133
5.4 Results………………………………………………………………………137
5.5 Discussion………………………………………………………………..…137
5.6 References…………………………………………………………………..141

vi

LIST OF PUBLICATIONS

Published:

1. Lynch, K.M., Shi, Y., Toga, A.W., Clark, K.A., 2018. Hippocampal Shape
Maturation in Childhood and Adolescence. Cerebral Cortex.
doi:10.1093/cercor/bhy244.

2. Sepehrband, F., Lynch, K.M., Cabeen, R.P., Gonzalez-Zacarias, C., Zhao, L.,
D’Arcy, of M., Kesselman, C., Herting, M.M., Dinov, I.D., Toga, A.W., Clark,
K.A., 2018. Neuroanatomical morphometric characterization sex differences in
youth using statistical learning. Neuroimage 172, 217–227.

In Preparation:

1. Lynch, K.M., Cabeen, R.P., Toga, A.W., Clark, K.A. The magnitude and timing
of major white matter tract maturation from infancy through adolescence with
NODDI. In preparation

2. Lynch, K.M., Page, K.A., Toga, A.W., Clark, K.A. The influence of BMI on
hippocampal shape in children. In preparation

vii

LIST OF TABLES

Table Page
Table 1.1 Hippocampal strata and their cell types in humans…………………………...18
Table 2.1 Description of NODDI parameters in bilateral whole-tract analyses....……...49
Table 3.1 Hippocampal shape development study demographics by site……………….72
Table 3.2 Significant clusters related to age and sex using shape analysis……………...79
Table 3.3 Significant age-related clusters in males and females………………………...82
Table 4.1 Study demographics for the multi-site dataset………………………………113

viii

LIST OF FIGURES

Figure Page
Figure 1.1 Timeline of major events in brain development………………………………10
Figure 1.2 Age-related differences of myelin in human post-mortem white matter…….12
Figure 1.3 Developmental differences in peripheral nerve axonal features…………….16
Figure 1.4 Hippocampal subfield and strata organization in human tissue……………17
Figure 1.5 Tissue compartments modeled with NODDI………………………………..23
Figure 2.1 Major white matter tracts visualized using tractography…………………….46
Figure 2.2 Age-related NDI increases in major white matter tracts………………….…50
Figure 2.3 Along-tract changes in NDI and ODI within the forceps minor…………….51
Figure 2.4 Along-tract changes in NDI and ODI within the corticospinal tract………...52
Figure 3.1 Age distribution of participants included in the present study………………73
Figure 3.2 Shape analysis method and hippocampal surface anatomy………………….75
Figure 3.3 Age-related changes to adjusted hippocampal volume stratified by sex…….78
Figure 3.4 Linear and non-linear hippocampal expansion with age…………………….81
Figure 3.5 Main effect of sex and age*sex interaction on hippocampal shape…………83
Figure 3.6 Maturational trajectory of left hippocampal surface in males and females….85
Figure 4.1 Regional hippocampal shape changes with BMI z-score…………………..116
Figure 5.1 Partial volume issues with sampling microstructure at the zero level set….135
Figure 5.2 Generation of inner iso-surface for parameter interpolation……………….136
Figure 5.3 Hippocampal microstructure maps…………………………………………137

9

Chapter 1
Introduction and Background

1.1 Structural brain development

Brain development relies on several complex and interrelated mechanisms, including
synaptic reorganization of gray matter regions and myelination of white matter connections
between those regions. Given the dependency of information transfer on the computational
capacity of neuronal units (Passingham et al. 2002), structural reorganization during development
may play a role in shaping the efficiency and specialization of cognitive functions, such as
memory, visuospatial attention, and language (Casey et al. 2000; Rice and Barone 2000; Ramsaran
et al. 2019). Child development is the global consequence of evolutionary, experiential, and
genetic influences, however the specific mechanisms are unknown (Jiang and Nardelli 2015).
Brain maturation is a complex and protracted process of progressive and regressive events
that plays a critical role in sensorimotor and cognitive development. The neuro-ontogenic process
begins during gestation and continues during postnatal development and follows a dynamic,
sequential, and temporally overlapping series of cellular events (Tau and Peterson 2010) (Figure
1.1). Following embryonic neuronal migration the subsequent establishment of connectivity,
fundamental changes in synapse maturation, stabilization, and elimination due to environmental
interactions drive the formation of functionally specialized and efficient networks (Jiang and
Nardelli 2015).
Synapse and neuronal overproduction continues until approximately 2 years of age, while
the systematic elimination of aberrant and redundant synapses and connections continues through
adolescence by which time nearly 50% of the infant synapses are eliminated (Huttenlocher and
Dabholkar 1997b; Kolb and Whishaw 2003). Axonal myelination is the last stage of white matter
development and is a critical process for the synchronized and rapid flow of neural impulses
through the brain. Neuronal and synaptic pruning, myelination, and axon caliber enlargement are
global and prolonged developmental phenomena that continue through childhood and adolescence
(Schröder et al. 1978; Stiles and Jernigan 2010; Dubois et al. 2014). Since progressive
developmental events, such as axon growth and myelination, co-occur with regressive events, such

10

as axonal and synaptic pruning, it is therefore critical to develop tools that allow researchers to
measure specific developmental processes in vivo.

Figure 1.1 Timeline of major events in brain development.
The diagram represents brain development commencing with neurulation, and subsequently
followed with neuronal proliferation and migration prenatally. Apoptosis, synaptogenesis and
myelination continue during postnatal development. Image adapted from (Giedd et al. 1999).

White matter and the hippocampus are two brain structures that undergo dynamic patterns
of maturation that coincide with the emergence of diverse cognitive behaviors (Maguire et al.
2000; Gaser and Schlaug 2003; Draganski et al. 2006; Driemeyer et al. 2008). White matter fibers
provide fast information transfer between cortical and subcortical structures and myelination is
believed to be responsible for the opening and closure of critical developmental periods (Mcgee et
al. 2005; Schwab and Strittmatter 2014). The hippocampus consists of functionally distinct
subregions with unique cytoarchitectonic distributions (Duvernoy et al. 2013) and is critically
involved in learning and memory behaviors (Manns et al. 2003; Wais et al. 2006). While current
MRI methods have allowed researchers to successfully characterize the gross structural changes
in white matter and the hippocampus during postnatal development in vivo (Knickmeyer and
Gouttard 2008; Lebel et al. 2008; Uematsu et al. 2012; Wierenga et al. 2015) more advanced
techniques are required to discriminate diverse structural maturational processes. In the following
sections, the relevant cellular and cytoarchitectonic features of white matter fibers and the
hippocampal formation are described and the developmental trajectories of different cellular
features are presented. In vivo neuroimaging techniques sensitized to unique WM and hippocampal

11

tissue properties are described with a focus on models that detect specific microstructural features.
The purpose of the following experiments is to map the developmental trajectory of specific
cellular and geometric processes with the aim to more comprehensively quantify structural brain
development.

1.2 White matter cellular development

1.2.1 Myelination

White matter is composed of myelinated and unmyelinated axons that form short- and long-
range connections in the brain. Myelin is a modified plasma membrane derived from glial
oligodendrocytes that functions as an electrical insulator due to its lipid-rich content (Baumann
and Pham-Dinh 2001; Schmitt et al. 2015). Myelination is the process by which oligodendrocytes
wrap around axons while extruding cytoplasm to form tightly bound multi-layer membrane sheaths
interrupted with regular, discontinuous intervals, known as the Nodes of Ranvier, which contain
densely packed voltage gated sodium channels (Poliak and Peles 2003). These nodes function as
a mechanism for salutatory conduction because local circuits generated due to excitation cannot
flow through the high-resistance sheath and instead jumps from node to node, resulting in faster
action potential (AP) propagation compared to unmyelinated axons. Fast and efficient information
transfer across neural systems is critical for the integration and synchronization of developing
neural systems critical for normative function and behavior.
The initial extension of oligodendrocyte processes and subsequent myelination rely on a
number of molecular signals derived from both myelin and axon. A single oligodendrocyte is
responsible for the development and maintenance of several myelin processes on different axons,
making CNS myelination a complex and bidirectional process. Axon selection is poorly
understood, however current research suggests that contact initiation is driven by factors including
the myelin-derived integrins and their ligands, laminins (Yang et al. 2005), which work in concert
with the axon-derived nerve growth factor (NGF) and axonal tyrosine kinase TrkA receptors (Chan
et al. 2004). `Microtubules and microfilaments aid myelin process extension, where the inner
mesaxon wraps about the axon to form the growing multilamellar myelin sheath. It is believed that
lipid boats (Munro 2003) and myelin basic protein (MBP) (Coleman et al. 1982) play a role in

12

essential molecular transport to the developing mesaxon. Following initiation of myelination,
axonal contactin-associated protein (Caspr) and contactin interact with glial neurofascin 155 to
form the paranodal axo-glial junction, which coincides with node formation (Einheber et al. 1997;
Rios et al. 2000; Tait et al. 2000).
Myelin thickness is governed by a number of physical, molecular, and electrical
constraints. In general, small-diameter axons (<1 μm) are not myelinated (Nave and Salzer 2006).
AP propagation in these small fibers would not experience significantly faster transmission rates
when myelinated due to the fractional space occupied by myelin and the resulting increase in axial
resistance (Chomiak and Hu 2009). Axons with larger diameters, however, share an interesting
relationship with myelin thickness. A common measure for the degree of myelination is the g-
ratio, or the ratio of the inner (axon) and outer (axon + myelin) cross-sectional diameters.
Theoretically, the optimal g-ratio for efficient information transfer is 0.6 (Rushton 1951), and
experimental evidence shows that the g-ratio range in most species is relatively stable throughout
the CNS with values ranging from 0.5 to 0.7 (Hildebrand and Hahn 1978; Berthold et al. 1983).

Figure 1.2 Age-related differences of myelin content in human post-mortem white matter
Association between age and (Left) myelinated fiber density and (right) myelin-associated
glycoprotein (MAG) density estimates in histological preparations of functionally-specialized
white matter tracts from post-mortem human tissue. Myelinated fiber density indicates the degree
of myelination and demonstrates that myelination continues well into the third decade of life. MAG
regulates interactions between neurons and oligodendrocytes and is involved in establishing adult-
like phenotypes in axon structure. Myelinated fiber density and MAG demonstrate regional
heterogeneity in white matter developmental trajectories. Adapted from (Miller et al. 2012b)

While the g-ratio remains relatively stable throughout the brain (Stikov et al. 2015), the
development of the processes that contribute to the g-ratio – myelin thickness and axon diameter

13

– undergo protracted developmental processes in human and non-human primates. Myelination is
characterized by a delayed period of maturation that extends beyond late adolescence and
continues into the third decade of life (Figure 1.2). Furthermore, white matter maturation is
asynchronous, with the timing of regional development depending on the hierarchical organization
of the connections (Guillery, 2005). Previous studies have shown that myelin develops earlier in
fibers that support primary cortical areas than fibers that support higher order cognitive behaviors
(Yakovlev and Lecours 1967; Lebel et al. 2008; Miller et al. 2012b). In vivo approaches sensitized
to developmental myelinating processes will contribute to knowledge regarding the emergence of
complex cognitive behaviors during development.

1.2.2 Axon diameter

After establishing their synaptic targets, axons increase diameter several fold during
postnatal development. Large diameter axons propagate electrical signals faster than small
diameter axons due to reduced membrane resistance, but nonetheless axon diameter varies widely
across the brain (~.1-10 um) (Perge et al. 2012) due to functional considerations and activity-
dependent mechanisms (Greenberg et al. 1990). Axon diameter is influenced by the cytoskeleton,
and includes neurofilaments, microtubules and actin (Leterrier et al. 2017). Neurofilaments are
neuron-specific intermediate filaments generically involved in the regulation of axon diameter
(Hsieh et al. 1994). Previous studies have demonstrated reduced axon diameters with slower
velocity in axons that have lost their neurofilaments (Ohara et al. 1993; Sakaguchi et al. 1993). A
potential mechanism for axon diameter regulation involve C-terminal domain neurofilament
phosphorylation, which causes expanded axon diameter (de Waegh et al. 1992) through radial
growth due to increased inter-filament spacing (Nixon et al. 1994).
Myelination heavily influences axon caliber as signals derived from myelinating
oligodendrocytes initiate radial axon growth (de Waegh et al. 1992). This is supported from
previous evidence demonstrating smaller diameters in unmyelinated axons associated with
decreased phosphorylation and denser neurofilaments (Hsieh et al. 1994). Myelinated axons have
larger diameters and, in particularly enlarged axons, the g-ratio appears to increase, resulting in a
relatively thinner myelin sheath (Berthold et al. 1983; Gillespie and Stein 1983; Chatzopoulou et
al. 2008). Furthermore, the relationship between axon diameter and myelination appears to be bi-

14

directional, as myelination is preferentially initiated in large diameter axons during development
(Almeida et al. 2011; Lee et al. 2012). Previous evidence has also established a role for neuronal
activity to influence axon diameter. In a recent electrophysiological experiment utilizing super-
resolution imaging, unmyelinated axons exhibit radial growth in response to high-frequency AP
firing ex vivo in a relatively short time frame (Chéreau et al. 2017). Progressive enlargement of
axon diameter in response to physiological activity is also demonstrated in the onset of
developmental processes. In the auditory brainstem, increases in myelinated axon diameter
coincide with functional acquisition of hearing (Sinclair et al. 2017). Experimental delays in
auditory stimuli are similarly accompanied by axon enlargement once sensory input is restored
(Sinclair et al. 2017). Together, these results show the interplay between neural activity,
myelination and axonal remodeling through changes in axon diameter.
While the precise timing of axon caliber maturation in humans is unknown, previous
studies in post-mortem human peripheral nerves show neurons reach their maximum diameter
between 4 and 5 years of age (Schröder et al. 1978; Friede and Beuche 1985; Jacobs and Love
1985). However, it is not clear if central nervous system (CNS) axon diameters mature at the same
rate as peripheral nerves because different criteria is required to commence myelination. For
example, CNS axons with axon diameters greater than 0.2 μm are myelinated, while peripheral
nervous system (PNS) axons myelinate after diameters reach 2 μm (Susuki 2010). If CNS axons
exhibit a similar maturational pattern of radial expansion as the PNS, then axon diameters would
reach mature phenotypes before myelinating processes.

1.2.3 Axon number

The gross white matter architecture is largely established by birth, however structural
connections are reorganized and refined throughout infancy and childhood. Evidence from human
and non-human primates show an overabundance synapses throughout the entire brain in the early
postnatal period (Huttenlocher and de Courten 1987; Zecevic et al. 1989; Bourgeois and Rakic
1993; Bourgeois et al. 1994; Huttenlocher and Dabholkar 1997a). This increased synaptogenesis
is attributed to a proliferation of dendrites, dendritic spines, and arborization of axon collaterals
(Huttenlocher 1979; Huttenlocher and Dabholkar 1997a). Major white matter tracts, such as the
corpus callosum, thalamocortical projections, and the corticospinal tract form transient

15

connections during the early postnatal period by rapidly sampling the local environment, forming
synapses, and retracting (Stanfield et al. 1982; Stanfield and O’Leary 1985a; Innocenti and Price
2005). Following the overproduction of connections, neurons undergo pruning through branch
retraction and synapse elimination to remove redundant connections and increase network
efficiency. Axons can undergo small-scale pruning at the level of the synapse (Yuste and
Bonhoeffer 2001; Bonhoeffer and Yuste 2002; Ethell and Pasquale 2005), and large-scale pruning
of long axon collaterals (O’Leary 1987; O’Leary et al. 1990; Luo and O’Leary 2005).
Small-scale pruning does not change the number of axons as these modifications occur
primarily on dendritic processes. Large-scale axon eliminations allow neurons to remove
redundant collaterall innervation and involves the loss of the distal part of the primary axon.
Elimination of large portions of axons have been observed in the vertebrate CNS during the early
postnatal period involving restructuring of visual callosal projections(Innocenti 1981; Aggoun‐
Zouaoui and Innocenti 1994; Houzel et al. 1994) and layer V sensory projections to subcortical
structures (Stanfield et al. 1982; Stanfield and O’Leary 1985a, 1985b; O’Leary 1987; O’Leary et
al. 1990). Post-mortem evidence in humans suggest that excessive and aberrant connections
decline to adult levels across childhood and adolescence (Bourgeois and Rakic 1993; Huttenlocher
and Dabholkar 1997a), which may coincide with cognitive and behavioral milestones (Gogtay et
al. 2004). New neurons are also formed in the postnatal brain, however neurogenesis does not
make an appreciable contribution to the overall axon number (Cayre et al. 2009).

These cellular components provide the foundation for white matter structure and exhibit
heterochronous change postnatally from childhood through adolescence. Axon caliber reaches
adult phenotypes in childhood, while myelination follows a protracted process through young
adulthood (Figure 1.3). The magnitude and timing of axonal pruning is less well understood.
Functionally normative development is dependent on the proper maturational timing for each of
these units as a consequence of experience-dependent network refinement. However, the study of
white matter development is complicated by its invasive nature and studies are limited to rodents,
primates and post-mortem human tissue (Stiles and Jernigan 2010). There is still uncertainty about
the temporal extent of white matter maturation during the postnatal period in humans. In vivo
approaches sensitized to specific cellular and tissue properties can help to disentangle these
different and important contributions to structural development.

16

Figure 1.3 Developmental differences in peripheral nerve axonal features
Electron microscopy of semithin slabs of ulnar nerve cross-sections at 13.7 (left) and 203.4 (right)
months of age in humans. White matter development in childhood is characterized by a progressive
structural changes including increased axon caliber, thicker myelin, and pruned axons. Figure
adapted from (Schröder et al. 1978)

1.3 Hippocampal cellular development

1.3.1 Hippocampal structure

The hippocampus is a complex subcortical structure consisting of both gray and white
matter within the entorhinal cortex. The hippocampus is critically involved in episodic and
semantic memory consolidation (Manns et al. 2003; Wais et al. 2006), working memory
(Ranganath and Blumenfeld 2005; Axmacher et al. 2007) and a broad range of non-mnemonic
functions, including attention and language processes (Shohamy and Turk-Browne 2013).
Hippocampal anatomy is a bilaminar structure composed of functionally distinct subfields with
different cytoarchitectonic distributions and consists of the subiculum, cornu ammonis (CA) 1-3,
hilar region (also known as CA4) and dentate gyrus (DG) (Duvernoy et al. 2013) (Figure 1.4).
The hippocampus is also contains cell-specific strata arranged transverse to the
hippocampal axis, with individual subfield lamina acting as a specific computational unit
(Blackstad et al. 1970; Andersen et al. 1971; Hjorth-Simonsen and Jeune 1972). The hippocampal
strata starting from the ventricular cavity to the vestigial hippocampal sulcus are described in
Table 1.1. The alveus, stratum oriens, stratum pyramidale, stratum lucidum (only in CA3), stratum

17

radiatum, stratum lacunosum and stratum moleculare are within the CA subfields, while the DG
contains the stratum moleculare, stratum granulosum, and polymorphic layer. The DG and CA
subfields are separated by the vestigial hippocampal sulcus (Insausti et al. 2010; Duvernoy et al.
2013). Neurogenesis is a postnatal developmental feature within the DG, where cells in the
subgranular zone (SGZ) adjacent to the stratum granulosum produce new neurons through
adulthood (Eriksson et al. 1998).

Figure 1.4 Hippocampal subfield and strata organization in human post-mortem tissue
Coronal slice of hippocampal anatomy in (left) a post-mortem photomicrograph from a five year
old child and (right) a schematic of the subfield and strata organization at the same level. The
hippocampus is composed of the CA1–CA4, subfields, DG, and the following strata from distal to
proximal for Cornu Ammonis: 1 alveus, 2 stratum oriens, 3 stratum pyramidale, 3’ stratum
lucidum, 4 stratum radiatum, 5 stratum lacunosum, 6 stratum moleculare, 7 vestigial hippocampal
sulcus; strata for the DG: 8 stratum moleculare, 9 stratum granulosum, 10 polymorphic layer, 11
fimbria, 12 margo denticulatus, 13 fimbriodentate sulcus, 14 superficial hippocampal sulcus, 15
subiculum, 16 choroid plexuses, 17 tail of caudate nucleus, 18 temporal (inferior) horn of the
lateral ventricle. Schematic adapted from (Duvernoy et al. 2013) and histology from (Insausti et
al. 2010)

The intra-hippocampal circuitry is composed of a network of serial and recurrent
connections between hippocampal subfields that are critical for learning and memory function
(Lisman 1999). This white matter fibers can be divided into two pathways: the polysynaptic
pathway, which consists of a relay of 3 synapses, and the direct pathway, which directly projects
to output hippocampal output neurons (Amaral and Insausti 1990; Witter and Groenewegen 1992;
Eichenbaum et al. 1994; Leonard and Squire 1995; Markowitsch 1995). The polysynaptic pathway

18

involves a 3-part relay: (1) the perforant pathway involves projections from the entorhinal cortex
(ERC) to the DG by perforating the subiculum (Amaral and Insausti 1990): (2) Granule cell
neurons from the DG project to primarily CA3 in the stratum lucidum and some CA4 via mossy
fibers (Sun et al. 2017), and (3) CA3/CA4 neurons project to the apical dendrites of CA1 through
Schaffer collaterals (Treves 1995). CA1 pyramidal neurons then project back to the ERC and
provides outputs to the rest of the cortex through the fornix by way of the fimbria (Ramón y Cajal
1911). The direct pathway involves projections from the ERC directly to the CA1 through the
perforant path (Duvernoy et al. 2013).

Table 1.1. Hippocampal strata and their cell types in humans
Strata Cell types
Alveus Acellular, heavily myelinated
Stratum oriens Basal dendrites, cell-sparse
Stratum pyramidale Pyramidal cell bodies
Stratum lucidum Mossy fiber terminations only found in the CA3 subfield
Stratum radiatum Apical dendrites of stratum pyramidale
Stratum lacunosum Main dendrites and distal ramifications of CA3 pyramidal cells;
appears cell-sparse
Stratum moleculare (CA) Cell-sparse region with dendritic projections
Hippocampal sulcus Vestigial hippocampal sulcus
Stratum moleculare (DG) Cell-sparse region with dendritic projections
Stratum granulosum Densely packed small and round neurons devoid of myelin
Polymorphic layer Thin layer with axons that cross from granule cell neurons to CA4

The overall architecture of the hippocampus is largely established at birth and does not
change significantly during infancy and childhood (Insausti et al. 2010), however hippocampal
subfields and circuitry undergo dynamic and heterogeneous postnatal development. The following
sections describe histological studies that have uncovered specific patterns of maturation in
hippocampal structure.

19

1.3.2 Hippocampal subfield maturation

The most robust changes to hippocampal structure occur during the first 2 postnatal years
(Utsunomiya et al. 1999), however key sub-regions undergo protracted developmental trajectories
that that continue beyond childhood and adolescence (Gogtay et al. 2006; Ghetti et al. 2010; Ofen
et al. 2019). Hippocampal biology undergoes changes at the cellular and molecular level during
the early postnatal period, including alterations in gene and receptor expression profiles (Alberini
and Travaglia 2017) (Law et al. 2003a, 2003b; Eastwood et al. 2006; Travaglia et al. 2016),
synaptogenesis (Dumas and Foster 1995) and downscaling of silent synapses (Liao et al. 1999;
Petralia et al. 1999), which have been proposed to trigger the onset of hippocampal function
(Dumas and Rudy 2010). Previous studies demonstrate that differences in hippocampal structure
are associated with memory performance (Van Petten 2004; Lavenex and Banta Lavenex 2013;
Keresztes et al. 2018); therefore, a complete understanding of the development of hippocampal
sub-components will contribute to our knowledge of the normative acquisition of learning and
memory behaviors.

Dentate gyrus
In primates, the majority of DG neurons form prenatally (Nowakowski and Rakic 1981),
however there is some evidence in humans demonstrating limited neuronal production postnatally
(Seress 2001). Substantial age-related changes in hippocampal volume occur during the third
trimester in humans, and these changes are primarily attributed to increased DG volume including
thickening of the granule layer, molecular layer, and hilar region (Seress 2001; Lee et al. 2017).
At birth, the granule cell layer is the most visible layer of the DG, as granule cells neatly stack up
with between 6 and 10 cells thick, however newborns lack the sharpness observed in human
granule cells (Insausti et al. 2010).
The DG undergoes a protracted maturation compared to other subfields, particularly within
the granule cell layer (Seress 2001; Insausti et al. 2010), with structural maturation continuing well
beyond childhood (Lavenex and Banta Lavenex 2013). In non-human primates, DG volume
demonstrates small increases in the first few postnatal months and, during that period, nearly 30%
of the total number of granule cells are added postnatally to reach adult levels (Jabès et al. 2010;
Jabès and Nelson 2015). However, evidence in non-human primates show that DG volume

20

increases more than two-fold between 3 months and 7-13 years, after granule cells reach mature
levels (Lavenex et al. 2007). This increase is attributed to synaptogenesis and elaboration of
dendritic trees within the subfield (Lavenex et al. 2007). Furthermore, increases in the size of the
SGZ. Furthermore, the SGZ increases substantially over the first 6 postnatal months (Seress and
Ribak 1995a, 1995b; Seress 2001) and this enlargement is primarily due to the extension of basal
dendrites from granule cells (Seress and Mrzljak 1987).

CA3
The CA3 subfield is characterized by large pyramidal neurons and inclusion of the stratum
lucidum where mossy fibers terminate (Insausti et al. 2010). The postnatal maturational trajectory
for the CA3 is protracted and parallels that of the DG (Lavenex and Banta Lavenex 2013).
Volumetric increases in CA3 are more apparent in the portion of the subfield proximate to the DG
where mossy fiber enter compared to ERC inputs (Amaral et al. 2007; Kondo et al. 2008, 2009).
While the maturation of mossy fibers are not well understood, one study demonstrated scarce
staining of mossy fibers using Timm staining in the rhesus monkey demonstrate scarce staining
that increases by the third month characterized by specialized thorny excrescences accompanied
by increased stratum lucidum thickness (Seress and Ribak 1995a). Together, these results show
that CA3 development is coupled with DG development due to their shared mossy fiber
projections.

CA2, CA1, and Subiculum
Postnatal changes to the CA2, CA1 and subiculum appear to be limited compared to the
CA3 and DG (Insausti et al. 2010). In the CA1 and CA2, there is an overall increase in the size of
projection neurons within the stratum pyramidale over the first postnatal year in humans, which is
accompanied by an increase in neuropil due to axonal and dendritic branching (Insausti et al.
2010). Unlike the late postnatal development of mossy fibers (Seress and Ribak 1995a), some of
the perforant pathway fibers projecting from the ERC to CA1 are present by the sixth gestational
month, though they also demonstrate some limited postnatal maturation (Benes 1989). While there
is rapid increases in the subiculum during the first postnatal year in humans, the overall appearance
of the subiculum does not change appreciably (Insausti et al. 2010).

21

Hippocampal circuitry
The heterogeneous trajectories of subfield development may provide clues for the
functional acquisition of flexible learning and memory processes (Lee et al. 2017). The portion of
the perforant path that forms the monosynaptic pathway between the ERC and CA1 appear to be
established prenatally. The portion of the perforant path connecting the ERC to the DG, forming
the first relay for the trisynaptic pathway, demonstrates sparse connectivity before birth (Hevner
and Kinney 1996). The early development of monosynaptic pathways and later development of
polysynaptic pathways demonstrate the hierarchical maturation of hippocampal circuitry (Lavenex
and Banta Lavenex 2013). It is therefore proposed that developing memory abilities must rely on
the inflexible cortical learning of monosynaptic pathways during the development of basic
episodic memory (Lavenex and Banta Lavenex 2013; Gómez and Edgin 2016) such as the ability
to form transitive associations (Cuevas et al. 2006). The refinement of episodic memory
consolidation is observed during the preschool years, which may be accompanied by the delayed
maturation of the more trisynaptic circuit (Lavenex and Banta Lavenex 2013). The flexibility of
the trisynaptic pathway for the emergence of complex associative relationships during preschool
is supported by the unique contributions of the DG to pattern completion and the CA3 to pattern
separation (Jabès and Nelson 2015).

1.4 Diffusion neuroimaging methods for white matter

1.4.1 Diffusion tensor imaging (DTI)

Diffusion MRI (dMRI) is an imaging method that is sensitive to the displacement pattern
of water molecules undergoing diffusion and provides unique insight into characteristics of in vivo
white matter structure (Beaulieu 2009). The standard clinical dMRI technique is diffusion tensor
imaging (DTI), which models white matter microstructural integrity through quantification of the
preferred diffusion direction of water molecules, known as fractional anisotropy (FA). FA is a
useful indicator of major microstructural changes, such as brain damage, and correlates well with
cognitive ability (Saygin et al. 2013; Takeuchi et al. 2016) and development (Lebel et al. 2008).
However, the study of white matter development using DTI parameters is complicated by
model assumptions that fail to capture the complex structural organization of myelinated fibers.

22

First, crossing fibers present a challenge for DTI-based tractography and parameter estimates
because the tensor only models a single fiber orientation (Assaf and Pasternak 2008; Jones et al.
2013), when over 90% of white matter voxels contain crossing fibers (Jeurissen et al. 2013). It is
important to utilize methods that accounts for orientation differences to more accurately represent
the underlying white matter architecture. Second, DTI parameters are susceptible to signal
contamination from cerebrospinal fluid (CSF), resulting in altered estimates (Jones and Cercignani
2010). For example, researchers found the diffusion parameter axial diffusivity (AD) with CSF
contamination is significantly higher than AD corrected using a free water elimination approach
and demonstrate spurious correlations with age (Metzler-Baddeley et al. 2012). While DTI
provides sensitivity to tissue microstructural changes, it is inherently non-specific. For example,
FA changes may be caused by local fluctuations in neurite density, orientation distribution, axon
radius distribution, and cell permeability (Beaulieu 2009), and this variability limits the biological
interpretation.

1.4.2 Neurite orientation dispersion and density imaging (NODDI)

Neurite orientation dispersion and density imaging (NODDI) is a model based on multi-shell
diffusion data acquired at multiple gradient strengths to investigate tissue compartments (Zhang et
al. 2012). NODDI models 3 compartments based on their diffusion profiles: (1) free diffusion
occurs in CSF, (2) hindered diffusion occurs in the extracellular space, and (3) restricted diffusion
occurs within intracellular space, which corresponds to axonal and dendritic compartments,
collectively known as neurites (Figure 1.5). NODDI calculates two quantities, the neurite density
index (NDI) obtained from the intracellular volume fraction, and the orientation dispersion index
(ODI) that reflects the degree of angular variation and reflects geometric complexity. By modeling
neurites as zero-radius sticks, NODDI provides an orientationally invariant measure of relative
neurite density that provides far more specific information regarding tissue microstructure than the
tensor model alone and enables neurite density estimation even in places of high disorder such as
the cortical and subcortical gray matter (Zhang et al. 2012). Furthermore, optical staining of
autometallographic myelin demonstrates histological estimates of neurite density is highly
correlated with NDI derived from NODDI (Jespersen et al. 2010).

23

Figure 1.5 Tissue compartments modeled with NODDI
NODDI models three tissue compartments based on diffusion data acquired with multiple gradient
strengths. NODDI models compartments with free diffusion, such as CSF, to mitigate the issue of
partial voluming observed with DTI. Low gradient strengths are sensitized to linear diffusion
observed in compartments with hindered diffusion, such as the extracellular space. High gradient
strengths are sensitized to nonlinear diffusion observed in regions with restricted diffusion, such
as the intracellular compartments of dendrites and axons. Figure altered and adapted from (Miller
2014)

The present research aims to model tissue microstructure of dendritic and axonal
morphology across child development using NODDI. This will provide major new insights into
the developmental processes that contribute to white matter structure because experience-
dependent structural remodeling during critical developmental periods leads to changes in synaptic
plasticity, such as axonal and synaptic sprouting, synaptogenesis, and dendritic arborization. Direct
measurements of these features would provide much greater insight into mechanisms governing
these developmental changes. This method will thus provide the unique opportunity to measure
microstructural parameters during development that were otherwise only restricted to post-mortem
histology.

24

1.5 Structural neuroimaging methods for the hippocampus

1.5.1 Volumetric approaches

T1-weighted (T1w) images are a basic pulse sequence in MR imaging sensitive to the T1
relaxation times of different tissues. T1w depends on the longitudinal relaxation of a tissue’s net
magnetization vector. Fatty tissue, such as myelinated axons, realigns quickly with the main
magnetic field, and its contrast appears bright. Water, on the other hand, slowly realigns with the
main magnetic field and appears darker. Due to varying longitudinal relaxation time of different
tissues, T1-weighted images are ideal for detecting tissue borders and are used in clinical research
for quantifying gross anatomical features of the brain, including cortical thickness and subcortical
volume (Greve and Fischl 2009).
Previous in vivo MRI studies of hippocampal volumetric development, yield mixed results
with some studies showing no age-related changes in hippocampal volume (Yurgelun-Todd et al.
2003; Gogtay et al. 2006; Knickmeyer and Gouttard 2008; Lin et al. 2013), while others have
found subtle increases in volume (Giedd et al. 1996; Uematsu et al. 2012; DeMaster et al. 2014).
Therefore, it is likely that whole-hippocampus volumetry captures a mixture of spatially- and
functionally-distinct developmental processes.
Hippocampal subfield segmentation techniques for structural MRI have been developed
recently for the study of individual subfield volumes (Winterburn et al. 2013; Iglesias et al. 2015;
Yushkevich, Pluta, et al. 2015). However, large cross-sectional studies of subfield volumetric
changes across the lifespan are not in agreement regarding the maturational trajectories of
individual subfields (Daugherty et al. 2016, 2017). Differing results from developmental studies
on hippocampal subfields may be due to technical limitations on the spatial resolution required to
resolve individual subfield boundaries in clinically feasible scans (Amunts et al. 2005; Wisse et
al. 2012; Winterburn et al. 2013).

25

1.5.2 Shape analysis

One analytical approach that overcomes the issues observed with subfield and whole
hippocampal volumetry is shape analysis. Shape analysis enables the study of morphological
surface features by converting anatomical volumes into a surface representation partitioned into
vertices that form a triangulated mesh (Shi, Lai, et al. 2014). This enables the fine-grain analysis
of regionally specific surface deformations that may better reflect local structural changes in
development. Elongated surfaces, such as the hippocampus, can be quantified with radial distance
measures, which represent local indices of hippocampal cross-sectional thickness (Thompson et
al. 2004; Shi et al. 2009). Previous hippocampal shape analysis studies have demonstrated spatial
patterns of surface morphology associated with neurodegenerative disorder (Costafreda et al.
2011), visual impairment (Leporé et al. 2009), heritable traits (Sabuncu et al. 2016) and
hippocampal-dependent behaviors (Voineskos et al. 2015). Shape analysis will therefore provide
enhanced sensitivity to the dynamic patterns of hippocampal subfield and strata development that
is commonly obscured with volumetric approaches.

1.6 Summary and overview of chapters

The goal of this dissertation is to characterize regional developmental patterns of white
matter and hippocampal structure from childhood through adolescence using in vivo MRI methods
sensitized to specific cellular features. The magnitude and timing of major white matter tract
maturation using NODDI is described from infancy through adolescence (Chapter 2). White
matter fibers undergo protracted NDI maturation and NDI developmental rates vary along the
lengths of individual tracts. Next, hippocampal development is characterized from infancy through
adolescence in a large multi-study cohort using shape analysis (Chapter 3). Hippocampal shape
undergoes nonlinear and regionally specific patterns of surface expansion and demonstrates
localized sex differences. Localized patterns of hippocampal atrophy were also associated with
adiposity measures in typically developing children, demonstrating potential hippocampal
structural vulnerabilities in response to childhood obesity (Chapter 4). Shape analysis is
inherently non-specific, however, and a given change in hippocampal shape may reflect a number
of structural phenomena within the underlying tissue. Therefore, in order to elucidate and better

26

understand the cellular features that give rise to macroscopic changes to hippocampal size, I
propose a novel analytical framework that combines the regional sensitivity afforded with shape
analysis and the microstructural sensitivity of diffusion models (Chapter 5). Together, these
experiments demonstrate the dynamic structural development of hippocampal and white matter
processes during childhood and adolescence and offer a methodological approach to more
precisely map microstructural features that reflect biological properties.

1.7 References

Aggoun‐ Zouaoui D, Innocenti GM. 1994. Juvenile Visual Callosal Axons in Kittens Display
Origin‐ and Fate‐ related Morphology and Distribution of Arbors. Eur J Neurosci. 6:1846–
1863.
Alberini XCM, Travaglia A. 2017. Infantile Amnesia : A Critical Period of Learning to Learn
and Remember. J Neurosci. 37:5783–5795.
Almeida RG, Czopka T, Ffrench-Constant C, Lyons DA. 2011. Individual axons regulate the
myelinating potential of single oligodendrocytes in vivo. Development. 138:4443–4450.
Amaral D, Insausti R. 1990. Hippocampal formation. In: Praxinos G, editor. The human nervous
system. San Diego: Academic Press. p. 711–755.
Amaral DG, Scharfman HE, Lavenex P. 2007. The dentate gyrus: fundamental neuroanatomical
organization (dentate gyrus for dummies). Prog Brain Res. 163.
Amunts K, Kedo O, Kindler M, Pieperhoff P, Mohlberg H, Shah NJ, Habel U, Schneider F,
Zilles K. 2005. Cytoarchitectonic mapping of the human amygdala, hippocampal region and
entorhinal cortex: Intersubject variability and probability maps. Anat Embryol (Berl).
210:343–352.
Andersen P, Bliss T, Skrede K. 1971. Lamellar organization of hippocampal excitatory
pathways. Exp Brain Res. 13:222–238.
Assaf Y, Pasternak O. 2008. Diffusion Tensor Imaging (DTI)-based White Matter Mapping in
Brain Research: A Review. J Mol Neurosci. 34:51–61.
Axmacher N, Mormann F, Cohen MX, Elger CE. 2007. Sustained Neural Activity Patterns
during Working Memory in the Human Medial Temporal Lobe. J Neurosci. 27:7807–7816.

27

Baumann N, Pham-Dinh D. 2001. Biology of oligodendrocyte and myelin in the mammalian
central nervous system. Physiol Rev. 81:871–927.
Beaulieu C. 2009. The biological basis of diffusion anisotropy. In: In: Johansen-Berg, H.,
Behrens TEJ (Eds. ., editor. Diffusion MRI: From Quantitative Measurement to In-vivo
Neuroanatomy. Academic Press. p. 105–129.
Benes F. 1989. Myelination of cortical-hippocampal relays during late adolescence. Schizophr
Bull. 15:585–593.
Berthold CH, Nilsson I, Rydmark M. 1983. Axon diameter and myelin sheath thickness in nerve
fibres of the ventral spinal root of the seventh lumbar nerve of the adult and developing cat.
J Anat. 136:483–508.
Blackstad TW, Brink K, Hem J, Jeune B. 1970. Distribution o f Hippocampal Mossy Fibers in
the Rat. An Experimental Study with Silver Impregnation Methods. J Comp Neurol.
138:433–449.
Bonhoeffer T, Yuste R. 2002. Spine motility: phenomenology, mechanisms and function.
Neuron. 35:1019–1027.
Bourgeois JP, Goldman-Rakic PS, Rakic P. 1994. Synaptogenesis in the prefrontal cortex of
rhesus monkeys. Cereb Cortex. 4:78–96.
Bourgeois JP, Rakic P. 1993. Changes of synaptic density in the primary visual cortex of the
macaque monkey from fetal to adult stage. J Neurosci. 13:2801–2820.
Casey BJ, Giedd JN, Thomas KM. 2000. Structural and functional brain development and its
relation to cognitive development. Biol Psychol. 54:241–257.
Cayre M, Canoll P, Goldman JE. 2009. Cell migration in the normal and pathological postnatal
mammalian brain. Prog Neurobiol. 88:41–63.
Chan JR, Watkins T a, Cosgaya JM, Zhang C, Reichardt LF, Shooter EM, Barres B a. 2004.
NGF controls axonal receptivity to myelination by Schwann cells or oligodendrocytes.
Neuron. 43:183–191.
Chatzopoulou E, Miguez A, Savvaki M, Levasseur G, Muzerelle A, Muriel M-P, Goureau O,
Watanabe K, Goutebroze L, Gaspar P, Zalc B, Karagogeos D, Thomas J-L. 2008. Structural
requirement of TAG-1 for retinal ganglion cell axons and myelin in the mouse optic nerve. J
Neurosci. 28:7624–7636.

28

Chéreau R, Saraceno GE, Angibaud J, Cattaert D, Nägerl UV. 2017. Superresolution imaging
reveals activity-dependent plasticity of axon morphology linked to changes in action
potential conduction velocity. Proc Natl Acad Sci U S A. 114:1401–1406.
Chomiak T, Hu B. 2009. What is the optimal value of the g-ratio for myelinated fibers in the rat
CNS? A theoretical approach. PLoS One. 4.
Coleman D, Kreibich G, Frey A, Sabatini D. 1982. Synthesis and incorporation of myelin
polypeptides into CNS myelin. J Cell Biol. 95:598–608.
Costafreda SG, Dinov ID, Tu Z, Shi Y, Liu CY, Kloszewska I, Mecocci P, Soininen H, Tsolaki
M, Vellas B, Wahlund LO, Spenger C, Toga AW, Lovestone S, Simmons A. 2011.
Automated hippocampal shape analysis predicts the onset of dementia in mild cognitive
impairment. Neuroimage. 56:212–219.
Cuevas K, Rovee-Collier C, Learmonth AE. 2006. Infants form associations between memory
representations of stimuli that are absent. Psychol Sci. 17:543–549.
Daugherty A, Flinn R, Ofen N. 2017. Hippocampal CA3-dentate gyrus volume uniquely linked
to improvement in associative memory from childhood to adulthood. Neuroimage. 153:75–
85.
Daugherty AM, Bender AR, Raz N, Ofen N. 2016. Age differences in hippocampal subfield
volumes from childhood to late adulthood. Hippocampus. 26:220–228.
de Waegh SM, Lee VMY, Brady ST. 1992. Local modulation of neurofilament phosphorylation,
axonal caliber, and slow axonal transport by myelinating Schwann cells. Cell. 68:451–463.
DeMaster D, Pathman T, Lee JK, Ghetti S. 2014. Structural development of the hippocampus
and episodic memory: Developmental differences along the anterior/posterior axis. Cereb
Cortex. 24:3036–3045.
Draganski B, Gaser C, Kempermann G, Kuhn H, Winkler J, Büchel C, May A. 2006. Temporal
and spatial dynamics of brain structure changes during extensive learning. J Neurosci.
26:6314–6317.
Driemeyer J, Boyke J, Gaser C, Buchel C, May A. 2008. Changes in gray matter induced by
learning—revisited. PLoS One. 3:e2669.
Dubois J, Kulikova S, Poupon C, Huppi PS, Hertz-Pannier L. 2014. The early development of
brain white matter: A review of imaging studies in fetuses, newborns, and infants.
Neuroscience. 276:48–71.

29

Dumas T, Rudy J. 2010. Development of the hippocampal memory system: Creating networks
and modifiable synapses. In: Blumberg M,, Freeman J,, Robinson S, editors. Oxford library
of neuroscience. Oxford handbook of developmental behavioral neuroscience. New York,
NY, US: Oxford University Press. p. 587–606.
Dumas TC, Foster TC. 1995. Developmental Increase in CA3-CA1 Hippocampal Slice
Presynaptic Function in the. J Neurophysiol. 73:1821–1828.
Duvernoy H, Cattin F, Risold P-Y. 2013. The Human Hippocampus: Functional anatomy,
vascularization and serial sections with MRI. 4th Ed. ed. Berlin, Germany: Springer.
Eastwood SL, Weickert CS, Webster MJ, Herman MM, Kleinman JE, Harrison PJ. 2006.
Synaptophysin Protein and mRNA Expression in the Human Hippocampal Formation From
Birth to Old Age. Hippocampus. 16:645–654.
Eichenbaum H, Otto T, Cohen NJ. 1994. Two functional components of the hippocampal
memory system. Behav Brain Sci. 17:449–518.
Einheber S, Zanazzi G, Ching W, Scherer S, Milner T a., Peles E, Salzer JL. 1997. The axonal
membrane protein Caspr, a homologue of neurexin IV, is a component of the septate-like
paranodal junctions that assemble during myelination. J Cell Biol. 139:1495–1506.
Eriksson P, Perfilieva E, Bjork-Eriksson T, Aborn A, Nordborg C, Peterson D, Gage F. 1998.
Neurogenesis in the adult human hippocampus. Nat Med. 4:1313–1317.
Ethell IM, Pasquale EB. 2005. Molecular mechanisms of dendritic spine development and
remodeling. Prog Neurobiol. 75:161–205.
Friede RL, Beuche W. 1985. Combined scatter diagrams of sheath thickness and fibre calibre in
human sural nerves: Changes with age and neuropathy. J Neurol Neurosurg Psychiatry.
48:749–756.
Gaser C, Schlaug G. 2003. Brain structures differ between musicians and non-musicians. J
Neurosci. 23:9240–9245.
Ghetti S, DeMaster DM, Yonelinas AP, Bunge SA. 2010. Developmental Differences in Medial
Temporal Lobe Function During Memory Encoding. J Neurosci. 30:9548–9556.
Giedd J, Blumenthal J, Jeffries N, Castellanos F, Liu H, Zijdenbos A, Paus T, Evans A, Rapoport
J. 1999. Brain development during childhood and adolescence: a longitudinal MRI study.
Nat Neurosci. 2:861–863.

30

Giedd J, Vaituzis C, Hamburger S, Lange N, Rajapakse JC, Kaysen D, Vauss YC, Rapoport JL.
1996. Quantitative MRI of the Temoral Lobe, Amydala, and Hippocampus in Normal
Human Development: Ages 4-18 Years. J Comp Neurol. 366:223–230.
Gillespie MJ, Stein RB. 1983. The relationship between axon diameter, myelin thickness and
conduction velocity during atrophy of mammalian peripheral nerves. Brain Res. 259:41–56.
Gogtay N, Giedd JN, Lusk L, Hayashi KM, Greenstein D, Vaituzis AC, Iii TFN, Herman DH,
Clasen LS, Toga AW, Rapoport JL, Thompson PM. 2004. Dynamic mapping of human
cortical development during childhood through early adulthood. PNAS. 101:8174–8179.
Gogtay N, Nugent TF, Herman DH, Ordonez A, Greenstein D, Hyashi KM, Clasen L, Toga AW,
Giedd JN, Rapoport JL, Thompson PM. 2006. Dynamic Mapping of Normal Human
Hippocampal Development. Hippocampus. 16:664–672.
Gómez RL, Edgin JO. 2016. The extended trajectory of hippocampal development: Implications
for early memory development and disorder. Dev Cogn Neurosci. 18:57–69.
Greenberg M, Leitao C, Trogadis J, Stevens J. 1990. Irregular geometries in normal
unmyelinated axons: A 3D serial EM analysis. J Neurocytol. 19:978–988.
Greve DN, Fischl B. 2009. Accurate and robust brain image alignment using boundary-based
registration. Neuroimage. 48:63–72.
Hevner RF, Kinney HC. 1996. Reciprocal entorhinal-hippocampal connections established by
human fetal midgestation. J Comp Neurol. 372:384–394.
Hildebrand C, Hahn R. 1978. Relation between myelin sheath thickness and axon size in spinal
cord white matter of some vertebrate species. J Neurol Sci. 38:421–434.
Hjorth-Simonsen A, Jeune B. 1972. Origin and Termination of the Hippocampal Perforant Path
in the Rat Studied by Silver Impregnation. J Comp Neurol. 144:215–232.
Houzel J ‐ C, Milleret C, Innocenti G. 1994. Morphology of Callosal Axons Interconnecting
Areas 17 and 18 of the Cat. Eur J Neurosci. 6:898–917.
Hsieh S-T, Kidd GJ, Crawford TO, Xu Z, Lin W-M, Trapp BD, Cleveland DW, Griffin JW.
1994. Regional modulation of neurofilament organization by myelination in normal axons. J
Neurosci. 14:6392–6401.
Huttenlocher P, Dabholkar A. 1997a. Regional differences in synaptogenesis in human cerebral
cortex. J Comp Neurol. 178:167–178.

31

Huttenlocher PR. 1979. Synaptic density in human frontal cortex - Developmental changes and
effects of aging. Brain Res. 163:195–205.
Huttenlocher PR, Dabholkar AS. 1997b. Regional differences in synaptogenesis in human
cerebral cortex. J Comp Neurol. 387:167–178.
Huttenlocher PR, de Courten C. 1987. The development of synapses in striate cortex of man.
Hum Neurobiol. 6:1–9.
Iglesias JE, Augustinack JC, Nguyen K, Player CM, Player A, Wright M, Roy N, Frosch MP,
McKee AC, Wald LL, Fischl B, Van Leemput K. 2015. A computational atlas of the
hippocampal formation using ex vivo, ultra-high resolution MRI: Application to adaptive
segmentation of in vivo MRI. Neuroimage. 115:117–137.
Innocenti GM. 1981. Growth and reshaping of axons in the establishment of visual callosal
connections. Science (80- ). 212:824–827.
Innocenti GM, Price DJ. 2005. Exuberance in the development of cortical networks. Nat Rev
Neurosci. 6:955–965.
Insausti R, Cebada-Sánchez S, Marcos P. 2010. Postnatal development of the human
hippocampal formation. Adv Anat Embryol Cell Biol. 206:1–86.
Jabès A, Banta Lavenex P, Amaral DG, Lavenex P. 2010. Quantitative analysis of postnatal
neurogenesis and neuron number in the macaque monkey dentate gyrus. Eur J Neurosci.
31:273–285.
Jabès A, Nelson CA. 2015. 20 years after ‘“The ontogeny of human memory : A cognitive
neuroscience perspective,”’ where are we? Int J Behav Dev. 39:293–303.
Jacobs JM, Love S. 1985. Qualitative and quantitative morphology of human sural nerve at
different ages. Brain. 108:897–924.
Jespersen SN, Bjarkam CR, Nyengaard JR, Chakravarty MM, Hansen B, Vosegaard T,
Østergaard L, Yablonskiy D, Nielsen NC, Vestergaard-Poulsen P. 2010. Neurite density
from magnetic resonance diffusion measurements at ultrahigh field: Comparison with light
microscopy and electron microscopy. Neuroimage. 49:205–216.
Jeurissen B, Leemans A, Tournier J-D, Jones DK, Sijbers J. 2013. Investigating the prevalence
of complex fiber configurations in white matter tissue with diffusion magnetic resonance
imaging. Hum Brain Mapp. 34:2747–2766.

32

Jiang X, Nardelli J. 2015. Cellular and molecular introduction to brain development. Neurobiol
Dis. 92:3–17.
Jones DK, Cercignani M. 2010. Twenty-five pitfalls in the analysis of diffusion MRI data. NMR
Biomed. 23:803–820.
Jones DK, Knösche TR, Turner R. 2013. White matter integrity, fiber count, and other fallacies:
the do’s and don’ts of diffusion MRI. Neuroimage. 73:239–254.
Keresztes A, Ngo CT, Lindenberger U, Werkle-bergner M, Newcombe NS. 2018. Hippocampal
Maturation Drives Memory from Generalization to Specificity. Trends Cogn Sci. 22:676–
686.
Knickmeyer R, Gouttard S. 2008. A structural MRI study of human brain development from
birth to 2 years. J Neurosci. 28:12176–12182.
Kolb B, Whishaw IQ. 2003. Fundamentals of human neurophysiology. 5th ed. ed. New York:
Worth Publishers.
Kondo H, Lavenex P, Amaral DG. 2008. Intrinsic connections of the macaque monkey
hippocampal formation: I. Dentate gyrus. J Comp Neurol. 511:497–520.
Kondo H, Lavenex P, Amaral DG. 2009. Intrinsic connections of the macaque monkey
hippocampal formation: II. CA3 connections. J Comp Neurol. 515:349–377.
Lavenex P, Banta Lavenex P. 2013. Building hippocampal circuits to learn and remember:
Insights into the development of human memory. Behav Brain Res. 254:8–21.
Lavenex P, Banta Lavenex P, Amaral DG. 2007. Postnatal development of the primate
hippocampal formation. Dev Neurosci. 29:179–192.
Law A, Weickert C, Webster M, Herman M, Kleinman J, Harrison P. 2003a. Changesin NMDA
receptor subunit mRNA during development of the human hippocampus. Ann N Y Acad
Sci. 1003:426–430.
Law AJ, Weickert CS, Webster MJ, Herman MM, Kleinman JE, Harrison PJ. 2003b. Expression
of NMDA receptor NR1 , NR2A and NR2B subunit mRNAs during development of the
human hippocampal formation. Eur J Neurosci. 18:1197–1205.
Lebel C, Walker L, Leemans A, Phillips L, Beaulieu C. 2008. Microstructural maturation of the
human brain from childhood to adulthood. Neuroimage. 40:1044–1055.

33

Lee JK, Johnson EG, Ghetti S. 2017. The Hippocampus from cells to systems: Structure,
connectivity, and functional contributions to memory and flexible cognition. Cham,
Switzerland: Springer International Publishing.
Lee S, Leach MK, Redmond SA, Chong SYC, Mellon SH, Tuck SJ, Feng ZQ, Corey JM, Chan
JR. 2012. A culture system to study oligodendrocyte myelination processes using
engineered nanofibers. Nat Methods. 9:917–922.
Leonard W, Squire F. 1995. Transient Memory Impairment the Entorhinal Cortex in Monkeys
and with Bilateral Lesions of. J Neurosci. 15:5637–5659.
Leporé N, Shi Y, Lepore F, Fortin M, Voss P, Chou YY, Lord C, Lassonde M, Dinov ID, Toga
AW, Thompson PM. 2009. Pattern of hippocampal shape and volume differences in blind
subjects. Neuroimage. 46:949–957.
Leterrier C, Dubey P, Roy S. 2017. The nano-architecture of the axonal cytoskeleton. Nat Rev
Neurosci. 18:713–726.
Liao D, Zhang X, Brien RO, Ehlers MD, Huganir RL. 1999. Regulation of morphological
postsynaptic silent synapses in developing hippocampal neurons. Nat Neurosci. 2:37–43.
Lin M, Fwu PT, Buss C, Davis EP, Head K, Muftuler LT, Sandman CA, Su M-Y. 2013.
Developmental changes in hippocampal shape among preadolescent children. Int J Dev
Neurosci. 31:473–481.
Lisman JE. 1999. Relating Hippocampal Circuitry to Function. Neuron. 22:233–242.
Luo L, O’Leary DDM. 2005. Axon Retraction and Degeneration in Development and Disease.
Annu Rev Neurosci. 28:127–156.
Maguire EA, Gadian DG, Johnsrude IS, Good CD, Ashburner J, Frackowiak RSJ, Frith CD.
2000. Navigation-related structural change in the hippocampi of taxi drivers. Proc Natl
Acad Sci. 97:4398–4403.
Manns JR, Hopkins RO, Squire LR. 2003. Semantic memory and the human hippocampus.
Neuron. 38:127–133.
Markowitsch H. 1995. Which brain regions are critically involved in the retrieval of old episodic
memory? Brain Res Rev. 21:117–127.
Mcgee AW, Yang Y, Fischer QS, Daw NW, Stephen M. 2005. Experience-Driven Plasticity of
Visual Cortex Limited by Myelin and Nogo Receptor. Science (80- ). 309:2222–2226.

34

Metzler-Baddeley C, O’Sullivan MJ, Bells S, Pasternak O, Jones DK. 2012. How and how not to
correct for CSF-contamination in diffusion MRI. Neuroimage. 59:1394–1403.
Miller DJ, Duka T, Stimpson CD, Schapiro SJ, Baze WB, McArthur MJ, Fobbs AJ, Sousa
AMM, Sestan N, Wildman DE, Lipovich L, Kuzawa CW, Hof PR, Sherwood CC. 2012.
Prolonged myelination in human neocortical evolution. Proc Natl Acad Sci U S A.
109:16480–16485.
Miller KL. 2014. Diffusion MRI: Diffusion Acquisition: Pushing the Boundaries, Diffusion
MRI. Elsevier.
Nave K, Salzer J. 2006. Axonal regulation of myelination by neuregulin 1. Curr Opin Neurobiol.
16:492–500.
Nixon R, Paskevich P, Sihag R, Thayer C. 1994. Phosphorylation on carboxyl terminus domains
of neurofilament proteins in retinal ganglion cell neurons in vivo: influences on regional
neurofilament accumulation, interneurofilament spacing, and axon caliber. J Cell Biol.
126:1031–1046.
Nowakowski RS, Rakic P. 1981. The Time of Origin of Neurons in the Hippocampal Region of
the Rhesus Monkey. J Comp Neurol. 196:99–128.
O’Leary DD. 1987. Remodelling of early axonal projections through the selective elimination of
neurons and long axon collaterals. Ciba Found Symp. 126:113–142.
O’Leary DD, Bicknese AR, De Carlos JA, Heffner CD, Koester SE, Kutka LJ, Terashima T.
1990. Target selection by cortical axons: alternative mechanisms to establish axonal
connections in the developing brain. Cold Spring Harb Symp Quant Bio. 55:453–468.
Ofen N, Tang L, Yu Q, Johnson EL. 2019. Memory and the developing brain: From description
to explanation with innovation in methods. Dev Cogn Neurosci. 36:100613.
Ohara O, Gahara Y, Miyake T, Teraoka H, Kitamura T. 1993. Neurofilament deficiency in quail
caused by nonsense mutation in neurofilament-L gene. J Cell Biol. 121:387–395.
Passingham RE, Stephan KE, Kötter R. 2002. The anatomical basis of functional localization in
the cortex. Nat Rev Neurosci. 3:606–616.
Perge JA, Niven J, Mugnaini E, Balasubramian V, Sterling P. 2012. Why do axons differ in
caliber? J Neurosci. 32:626–638.

35

Petralia RS, Esteban JA, Wang Y, Partridge JG, Zhao H, Wenthold RJ, Malinow R, Spring C,
Harbor CS, York N. 1999. Selective acquisition of AMPA receptors over postnatal
development suggests a molecular basis for silent synapses. Nat Neurosci. 2:31–36.
Poliak S, Peles E. 2003. The local differentiation of myelinated axons at nodes of Ranvier. Nat
Rev Neurosci. 4:968–980.
Ramón y Cajal S. 1911. Histologie du systeme nerveux de l’homme et des vertbres. Paris:
Maloine.
Ramsaran AI, Schlichting ML, Frankland PW. 2019. The ontogeny of memory persistence and
specificity. Dev Cogn Neurosci. 36:100591.
Ranganath C, Blumenfeld RS. 2005. Doubts about double dissociations between short- and long-
term memory. TRENDS Cogn Sci. 9:374–380.
Rice D, Barone S. 2000. Critical periods of vulnerability for the developing nervous system:
Evidence from humans and animal models. Environ Health Perspect. 108:511–533.
Rios JC, Melendez-Vasquez C V, Einheber S, Lustig M, Grumet M, Hemperly J, Peles E, Salzer
JL. 2000. Contactin-associated protein (Caspr) and contactin form a complex that is targeted
to the paranodal junctions during myelination. J Neurosci. 20:8354–8364.
Rushton AH. 1951. A theory of the effects of fibre size in medullated nerve. J Physiol. 115:101–
122.
Sabuncu MR, Ge T, Holmes AJ, Smoller JW, Buckner RL, Fischl B, Alzheimer’s Disease
Neuroimaging Initiative the ADN. 2016. Morphometricity as a measure of the
neuroanatomical signature of a trait. Proc Natl Acad Sci U S A. 113:E5749-56.
Sakaguchi T, Okada M, Kitamura T, Kawasaki K. 1993. Reduced diameter and conduction
velocity of myelinated fibers in the sciatic nerve of a neurofilament-deficient mutant quail.
Neurosci Lett. 153:65–68.
Saygin ZM, Norton ES, Osher DE, Beach SD, Cyr AB, Ozernov-Palchik O, Yendiki A, Fischl B,
Gaab N, Gabrieli JDE. 2013. Tracking the roots of reading ability: white matter volume and
integrity correlate with phonological awareness in prereading and early-reading
kindergarten children. J Neurosci. 33:13251–13258.
Schmitt S, Castelvetri L, Simons M. 2015. Metabolism and functions of lipids in myelin.
Biochim Biophys Acta. 1851:999–1005.

36

Schröder JM, Bohl J, Brodda K. 1978. Changes of the ratio between myelin thickness and axon
diameter in the human developing sural nerve. Acta Neuropathol. 43:169–178.
Schwab ME, Strittmatter SM. 2014. Nogo limits neural plasticity and recovery from injury. Curr
Opin Neurobiol. 0:53–60.
Seress L. 2001. Morphological changes of the human hippocampal formation from midgestation
to early childhood. In: Nelson AA,, Luciana M, editors. Handbook of developmental
cognitive neuroscience. Cambridge, MA: MIT Press. p. 45–47.
Seress L, Mrzljak L. 1987. Basal dendrites of granule cells are normal features of the fetal and
adult dentate gyrus of both monkey and human hippocampal formations. Brain Res.
405:169–174.
Seress L, Ribak CE. 1995a. Postnatal development of CA3 pyramidal neurons and their afferents
in theammon’s horn of rhesus monkeys. Hippocampus. 5:217–231.
Seress L, Ribak CE. 1995b. Postnatal development and synaptic connections of hilar mossy cells
in the hippocampal dentate gyrus of rhesus monkeys. J Comp Neurol. 355:93–110.
Shi Y, Lai R, Wang DJJ, Pelletier D, Mohr D, Sicotte N, Toga AW. 2014. Metric optimization
for surface analysis in the Laplace-Beltrami embedding space. IEEE Trans Med Imaging.
33:1447–1463.
Shi Y, Morra J, Thompson P, Toga A. 2009. Inverse-consistent surface mapping with Laplace-
Beltrami Eigen-Features. Comp Imag Vis. 21:467–478.
Shohamy D, Turk-Browne NB. 2013. Mechanisms for widespread hippocampal involvement in
cognition. J Exp Psychol Gen. 142:1159–1170.
Sinclair JL, Fischl MJ, Alexandrova O, Heβ M, Grothe B, Leibold C, Kopp-Scheinpflug C.
2017. Sound-evoked activity influences myelination of brainstem axons in the trapezoid
body. J Neurosci. 37:8239–8255.
Stanfield BB, O’Leary DDM. 1985a. The transient corticospinal projection from the occipital
cortex during the postnatal development of the rat. J Comp Neurol. 238:236–248.
Stanfield BB, O’Leary DDM. 1985b. Fetal occipital cortical neurons transplanted to the rostral
cortex can extend and maintain a pyramidal tract axon. Nature. 313:135–137.
Stanfield BB, O’Leary DDM, Fricks C. 1982. Selective collateral elimination in early postnatal
development restricts cortical distribution of rat pyramidal tract neurones. Nature. 298:371–
373.

37

Stikov N, Campbell JSW, Stroh T, Lavel??e M, Frey S, Novek J, Nuara S, Ho MK, Bedell BJ,
Dougherty RF, Leppert IR, Boudreau M, Narayanan S, Duval T, Cohen-Adad J, Picard PA,
Gasecka A, C??t?? D, Pike GB. 2015. Quantitative analysis of the myelin g-ratio from
electron microscopy images of the macaque corpus callosum. Data Br. 4:368–373.
Stiles J, Jernigan TL. 2010. The basics of brain development. Neuropsychol Rev. 20:327–348.
Sun Y, Grieco SF, Holmes TC, Xu X. 2017. Local and Long-Range Circuit Connections to Hilar
Mossy Cells in the Dentate Gyrus. Integr Syst. 4:1–21.
Susuki K. 2010. Myelin: A specialized membrane for cell communication. Nat Educ. 3:59.
Tait S, Gunn-Moore F, Collinson JM, Huang J, Lubetzki C, Pedraza L, Sherman DL, Colman
DR, Brophy PJ. 2000. An oligodendrocyte cell adhesion molecule at the site of assembly of
the paranodal axo-glial junction. J Cell Biol. 150:657–666.
Takeuchi H, Taki Y, Hashizume H, Asano K, Asano M, Sassa Y, Yokota S, Kotozaki Y, Nouchi
R, Kawashima R. 2016. Impact of reading habit on white matter structure: Cross-sectional
and longitudinal analyses. Neuroimage. 133:378–389.
Tau GZ, Peterson BS. 2010. Normal development of brain circuits. Neuropsychopharmacology.
35:147–168.
Thompson PM, Hayashi KM, De Zubicaray GI, Janke AL, Rose SE, Semple J, Hong MS,
Herman DH, Gravano D, Doddrell DM, Toga AW. 2004. Mapping hippocampal and
ventricular change in Alzheimer disease. Neuroimage. 22:1754–1766.
Travaglia A, Bisaz R, Cruz E, Alberini CM. 2016. Neurobiology of Learning and Memory
Developmental changes in plasticity , synaptic , glia and connectivity protein levels in rat
dorsal hippocampus. Neurobiol Learn Mem. 135:125–138.
Treves A. 1995. Quantitative Estimate of the Information Relayed by the Schaffer Collaterals. J
Comput Neurosci. 2:259–272.
Uematsu A, Matsui M, Tanaka C, Takahashi T, Noguchi K, Suzuki M, Nishijo H. 2012.
Developmental Trajectories of Amygdala and Hippocampus from Infancy to Early
Adulthood in Healthy Individuals. PLoS One. 7:e46970.
Utsunomiya H, Takano K, Okazaki M, Mitsudome A. 1999. Development of the temporal lobe
in infants and children: Analysis by MR-based volumetry. Am J Neuroradiol. 20:717–723.

38

Van Petten C. 2004. Relationship between hippocampal volume and memory ability in healthy
individuals across the lifespan: Review and meta-analysis. Neuropsychologia. 42:1394–
1413.
Voineskos AN, Winterburn JL, Felsky D, Pipitone J, Rajji TK, Mulsant BH, Chakravarty MM.
2015. Hippocampal (subfield) volume and shape in relation to cognitive performance across
the adult lifespan. Hum Brain Mapp. 36:3020–3037.
Wais PE, Wixted JT, Hopkins RO, Squire LR. 2006. The Hippocampus Supports both the
Recollection and the Familiarity Components of Recognition Memory. Neuron. 49:459–
466.
Wierenga LM, van den Heuvel MP, van Dijk S, Rijks Y, de Reus MA, Durston S. 2015. The
development of brain network architecture. Hum Brain Mapp. 37:717–729.
Winterburn JL, Pruessner JC, Chavez S, Schira MM, Lobaugh NJ, Voineskos AN, Chakravarty
MM. 2013. A novel in vivo atlas of human hippocampal subfields using high-resolution 3T
magnetic resonance imaging. Neuroimage. 74:254–265.
Wisse LEM, Gerritsen L, Zwanenburg JJM, Kuijf HJ, Luijten PR, Biessels GJ, Geerlings MI.
2012. Subfields of the hippocampal formation at 7T MRI: In vivo volumetric assessment.
Neuroimage. 61:1043–1049.
Witter M, Groenewegen H. 1992. Organizational principles of hippocampal connections. In:
Trimble M,, Bolwig T, editors. The temporal lobes and the limbic system. Petersfield:
Wrightson Biomedical. p. 37–60.
Yakovlev P, Lecours A. 1967. The myelogenetic cycles of regional maturation of the brain.
Minkowski. ed, Regional development of the brain in early life. Oxford: Blackwell.
Yang D, Bierman J, Tarumi YS, Zhong YP, Rangwala R, Proctor TM, Miyagoe-Suzuki Y,
Takeda S, Miner JH, Sherman LS, Gold BG, Patton BL. 2005. Coordinate control of axon
defasciculation and myelination by laminin-2 and -8. J Cell Biol. 168:655–666.
Yurgelun-Todd DA, Killgore WDS, Cintron CB. 2003. Cognitive correlates of medial temporal
lobe development across adolescence: A magnetic resonance imaging study. Percept Mot
Skills. 96:3–17.

39

Yushkevich PA, Pluta JB, Wang H, Xie L, Ding SL, Gertje EC, Mancuso L, Kliot D, Das SR,
Wolk DA. 2015. Automated volumetry and regional thickness analysis of hippocampal
subfields and medial temporal cortical structures in mild cognitive impairment. Hum Brain
Mapp. 36:258–287.
Yuste R, Bonhoeffer T. 2001. Morphological changes in dendritic spines associated with long-
term synaptic plasticity. Annu Rev Neurosci. 24:1071–1089.
Zecevic N, Bourgeois JP, Rakic P. 1989. Changes in synaptic density in motor cortex of rhesus
monkey during fetal and postnatal life. Dev Brain Res. 50:11–32.
Zhang H, Schneider T, Wheeler-Kingshott CA, Alexander DC. 2012. NODDI: practical in vivo
neurite orientation dispersion and density imaging of the human brain. Neuroimage.
61:1000–1016.

40

Chapter 2
Magnitude and timing of major white matter tract maturation from infancy through
adolescence with NODDI

2.1 Abstract

White matter maturation is a nonlinear and heterogeneous phenomenon characterized by
axonal packing, increased axon caliber, and a prolonged period of myelination. While current in
vivo diffusion MRI (dMRI) methods, like diffusion tensor imaging (DTI), have successfully
characterized the gross structure of major white matter tracts, these measures lack the specificity
required to unravel the distinct processes that contribute to microstructural development. Neurite
orientation dispersion and density imaging (NODDI) is a dMRI approach that probes tissue
compartments and provides biologically meaningful measures that quantify neurite density index
(NDI) and orientation dispersion index (ODI). The purpose of this study was to characterize the
magnitude and timing of major white matter tract maturation with NODDI from infancy through
adolescence in a cross-sectional cohort of 105 subjects (0.1-18.8 years). To probe the regional
nature of white matter development, we use an along-tract approach that partitions tracts to enable
more fine-grained analysis. Major white matter tracts showed exponential age-related changes in
NDI with spatially distinct maturational patterns. Overall, analyses revealed callosal fibers
developed before association fibers. Our along-tract analyses elucidate developmental hemispheric
asymmetries within the corticospinal tract (CST) and superior longitudinal fasciculus (SLF), which
may be reflective of their functionally specialized roles. ODI was not significantly associated with
age in the majority of tracts. Our results support the conclusion that white matter tract maturation
is heterochronous process and, furthermore, we demonstrate variability in the developmental
timing within major white matter tracts. Together, these results help to disentangle the distinct
processes that contribute to and more specifically define the time course of white matter
maturation.

41

2.2 Introduction

Human postnatal brain development is a complex process that follows a dynamic and
temporally overlapping series of cellular events that includes the development and formation of
axonal pathways through cellular proliferation, differentiation, synaptogenesis, apoptosis, and a
prolonged period of myelination (Rice and Barone 2000). Structural MRI enables the
quantification of gross anatomical changes during development that reflect these cellular
processes. Childhood and adolescence are marked by significant decreases in cortical gray matter,
indicative of synaptic pruning (Huttenlocher 1979; Gogtay et al. 2004), and increases in cerebral
white matter, believed to reflect myelination (Matsuzawa et al. 2001; Miller et al. 2012a).
Quantification of white matter microstructure in vivo is made possible with diffusion MRI
(dMRI), a technique that elucidates the underlying white matter organization by measuring water
displacement patterns in the brain (Beaulieu 2002). Diffusion tensor imaging (DTI) techniques
have been used to probe microstructural changes in major white matter tracts. Developmental
studies in children and adolescents using DTI have consistently demonstrated heterogeneous and
nonlinear increases in fractional anisotropy (FA) with age (Eluvathingal et al. 2007; Lebel et al.
2008, 2010, 2012; Tamnes et al. 2010; Imperati et al. 2011; Lebel and Beaulieu 2011; Clayden et
al. 2012; Wierenga et al. 2015; Krogsrud et al. 2016; Ullman and Klingberg 2016). While DTI-
based metrics are sensitive to age-related changes in microstructure, they are not specific (Pierpaoli
and Basser 1996; Beaulieu 2009). For example, an increase in FA may be due to an increase in
neurite density, a decrease in orientation dispersion, or a combination of factors (Jones et al. 2013).
Additionally, DTI measures suffer from partial volume effects due to free-water contamination
and complex fiber orientations, which bias the diffusion-weighted signal complicate the
interpretation of changes in DTI parameters (Jones and Cercignani 2010).
Multi-compartment dMRI models overcome these limitations by acquiring diffusion-
weighted images at multiple gradient strengths in order to probe tissue compartments with different
diffusion profiles (Stanisz et al. 1997; Assaf and Cohen 2000). Neurite Orientation Dispersion and
Density Imaging (NODDI) is a multi-compartment model that estimates tissue compartment
parameters associated with free, hindered, and restricted diffusion (Zhang et al. 2012). NODDI
models the intracellular volume fraction, known as neurite density index (NDI), and orientation
dispersion index (ODI), which provide biologically meaningful measures that quantify specific

42

microstructural processes. NDI measures the amount of restricted diffusion within a voxel and is
considered an index of neurite density. ODI describes the degree of neurite angular variation,
which allows for modeling the full spectrum of neurite orientation patterns observed in the brain.
Previous studies of white matter development using NODDI demonstrate age-related
increases in NDI across tracts in childhood and adolescence (Jelescu et al. 2014; Chang et al. 2015;
Kodiweera et al. 2015; Mah et al. 2017; Geeraert et al. 2019) and NDI has been shown to account
for more age-related variance than DTI measures in development (Genc et al. 2017; Mah et al.
2017). Results on age-related changes in ODI are less conclusive, as some studies show increases
(Chang et al. 2015; Kodiweera et al. 2015) or no change (Mah et al. 2017) with age. Together,
these findings suggests that NODDI parameters exhibit enhanced sensitivity to developmental
processes in white matter compared to DTI.
The main limitation of calculating a global measure to characterize an entire white matter
tract is that it may obscure potentially valuable information about focal developmental changes
within the tract. Along-tract approaches provide a framework to ameliorate this limitation by
characterizing the spatial pattern of diffusion parameters within individual white matter bundles.
This tract-based approach also provides enhanced sensitivity to regional developmental
trajectories in major white matter tracts. Previous diffusion studies using along-tract approaches
found diffusion parameters and their developmental trajectories vary along the length of individual
fiber bundles in toddlers (Goodlett et al. 2009; Johnson et al. 2014) and children/adolescents
(Colby et al. 2011; Yeatman, Dougherty, Myall, et al. 2012; Chen et al. 2016). Therefore, it is
reasonable to expect that NODDI measures will also demonstrate distinct regional patterns of
development within white matter tracts but with the added benefit of providing enhanced
sensitivity to microstructural features. Here, we propose to characterize the magnitude and timing
of development reflected with NODDI parameters in major white matter tracts using an along-
tract approach put forth by Colby et al. (Colby et al. 2012). This approach first consistently
parameterizes streamlines representing each bundle to match to a reference bundle; then the bundle
is partitioned cross-sectionally into equidistant points that can be statistically analyzed across the
population. In contrast to whole bundle analysis, this can provide enhanced sensitivity for the
detection of heterochronous patterns of white matter development that may occur at finer
anatomical scales than otherwise available when examining parameters averaged across entire
bundles.

43

While some studies have explored the age-related NODDI parameter changes of major
white matter tracts in infancy (Jelescu et al. 2014; Kunz et al. 2014) and childhood/adolescence
(Chang et al. 2015; Genc et al. 2017; Mah et al. 2017; Geeraert et al. 2019), the present study is
the first to examine the complete developmental trajectory through infancy, childhood, and
adolescence. The purpose of this study was to characterize the magnitude and timing of major
white matter tract maturation with NODDI in a cross-sectional cohort of 104 participants (0.1-18.8
years). Furthermore, we utilize along-tract analyses to spatially characterize developmental
patterns along the length of major white matter tracts. The NODDI microstructural parameters
may provide unique insight into the complex development of white matter tracts during this
dynamic period.

2.3 Methods

Subjects

Neuroimaging and demographic data were obtained through the publicly accessible
Cincinnati MR Imaging of NeuroDevelopment (C-MIND) data repository
(http://research.cchmc.org/c-mind) at Cincinnati Children’s Hospital (Holland et al. 2015).
Participants had no self-reported history of head trauma or neurological, developmental or
psychiatric disease. In total, 105 right-handed typically developing children and adolescents
between the ages of 0.1 and 18.8 years were included in the cross-sectional study (M=7.8, SD=4.9
years; 56 female). There were no significant sex differences in the age distribution of the cohort,
t(103)=1.54, p=.13. Parent/guardian consent were obtained for all subjects and procedures were
approved by the Institutional Review Board of the Cincinnati Children’s Hospital Medical Center
(CCHMC).

Image acquisition

MRI data was obtained at CCHMC using a Philips 3T Achieva system and a 32-channel
head coil. Diffusion-weighted images (DWI) were obtained using a spin-echo, echo-planar
imaging (EPI) method with intravoxel incoherent motion (IVIM) gradients. For each subject, 2

44

DWIs were obtained with the following parameters: 2 mm isotropic voxels, 112x109 acquisition
matrix, and 61 noncollinear diffusion sensitizing gradient directions with 7 non-diffusion-
weighted (b0) images interspersed throughout the acquisition and averaged into a single image.
The scans differed in the following parameters: (1) TR = 6614 ms, TE=81 ms, b = 1000 s/mm
2

and (2) TR=8112 ms, TE=81 ms, b=3000 s/mm
2
. T1-weighted images were acquired for
anatomical alignment using turbo-field echo MPRAGE sequence with the following parameters:
1 mm isotropic voxels, 256x224x160 mm FOV, TI=939 ms, TR=8.1 ms, TE = 3.7 ms.

Data preprocessing

Quality assurance procedures, motion correction, and intensity normalization were
performed on each MRI image using the C-MIND pipeline. All MRI images were skull-stripped
using FSL’s BET (Battaglini et al. 2008). The NODDI model was estimated using multi-shell
diffusion data acquired with different acquisition parameters, which may lead to model estimation
errors. We mitigated the effects of unmatched diffusion protocols by normalizing the shells with
the B0 volumes, as recommended in previous retrospective studies (Owen et al. 2014; Kelly et al.
2016). Signal normalization was performed by first producing a pair of average baseline scans
from b0 images with matching TR and TE, then each diffusion weighted image was matched and
normalized to its corresponding average baseline. For each subject, the two DWIs were co-
registered to a common space with corrected gradient tables using AIR (Woods et al. 1998). For
each subject, T1-weighted images were registered to the aligned DWI using an affine
transformation with FSL’s FLIRT (Jenkinson et al. 2002).

Microstructure parameters

NODDI and DTI parameters were calculated per voxel across each participant’s brain in
native space. DTI parameters were included for template registration.
DTI Fractional anisotropy (FA), axial diffusivity (AD), radial diffusivity (RD) and mean
diffusivity (MD) were computed using a two-stage weighted-least squares estimation scheme
(Veraart et al. 2013).

45

NODDI NODDI parameters were estimated using in-house software based on the NODDI Matlab
Toolbox (Zhang et al. 2012).The NODDI framework models 3 separate compartments: (1) the
intracellular compartment, such as axons and dendrites, is modeled as zero-radius cylinders, (2)
the extracellular compartment, such as glia and cell bodies, is modeled with Gaussian anisotropic
diffusion, and (3) the cerebrospinal fluid (CSF) compartment is modeled with isotropic diffusion.
The full normalized signal is a mixture of these 3 diffusion environments:

𝑆 = (1 − 𝑣 𝑖𝑠𝑜 )(𝑣 𝑖𝑐
𝑆 𝑖𝑐
+ (1 − 𝑣 𝑖𝑐
)𝑆 𝑒𝑐
) + 𝑣 𝑖𝑠𝑜 𝑆 𝑖𝑠𝑜 (1)

where S is the full normalized signal; Sic, Sec, and Siso are the normalized signals of the intracellular,
extracellular, and CSF compartments, respectively; and vic and viso are the normalized tissue
density estimates of the intracellular and CSF compartments, respectively (Zhang et al. 2012). The
orientation dispersion index (ODI), which describes the extent that the FOD deviates from its
primary diffusion direction, was modeled using the Watson distribution (Zhang, Hubbard, et al.
2011).

Fiber tracking

An atlas-based streamline tractography approach was used to obtain fiber bundle models
from the DWI data using the Quantitative Imaging Toolkit (QIT) (Cabeen et al. 2018) and other
software, noted where applicable. We first manually delineated reference fiber bundles in the IIT
ICBM diffusion MRI template (Zhang, Peng, et al. 2011; Varentsova et al. 2014) using
deterministic streaming tractography with a kernel regression framework (Cabeen et al. 2016).
These reference bundles were segmented according to a priori anatomical information from white
matter atlases and reference texts (Catani et al. 2002; Wakana et al. 2007; Catani and de Schotten
2012), and resulted in seed, inclusion, and exclusion region of interest (ROI) masks for each tract.
The diffusion tensor imaging data of each subject was used to compute a deformation from the
template to subject space using the tensor-based deformable registration algorithm in DTI-TK
(Zhang et al. 2006). The ROI masks for each tract were transformed to subject native space for
use in subject-specific bundle reconstruction. We used the ball-and-sticks model (Behrens et al.
2007) estimated using FSL bedpostx. We then reconstructed each bundle using the deterministic

46

streamline tractography with model-based Trilinear interpolation (Cabeen et al. 2016) with a
minimum volume fraction of 0.05, an angle threshold of 75 degrees, a step size of 0.5 mm, a
minimum length of 10 mm, and Runge-Kutta integration. For each bundle, seeding was initiated
from 5 randomly placed samples within the deformed seed ROI mask, tracks were retained using
the deformed inclusion and exclusion ROI masks, and a track density map was computed. The
following 9 major bilateral major white matter tracts were reconstructed for analysis: forceps major
(FMAJ), forceps minor (FMIN), anterior thalamic radiation (ATR), cingulum (CGC),
corticospinal tract (CST), inferior fronto-occipital fasciculus (IFO), inferior longitudinal fasciculus
(ILF), superior longitudinal fasciculus (SLF), and uncinated fasciculus (UNC) (Figure 2.1). All
tracts were visually inspected and edited for anatomical accuracy in native space.

Figure 2.1 Major white matter tracts used in the present study visualized using tractography
in a sample subject.
(Top) forceps minor (FMIN), cingulum (CGC), corticospinal tract (CST), inferior longitudinal
fasciculus (ILF) and forceps major (FMAJ). (Bottom) Anterior thalamic radiation (ATR),
uncinated fasciculus (UNC), superior longitudinal fasciculus (SLF), inferior fronto-occipital
fasciculus (IFO)

47

Along-tract Processing

To enable more fine-grained analysis of NODDI parameters within white matter tracts, we
employed an along-tract technique based on (Colby et al. 2012). This approach partitions each
tract cross-sectionally into equidistant points using an automated process, described as follows.
First, for each bundle in the template, a single prototypical curve was found by identifying the
“centroid” curve with the minimum Hausdorff distance to the others, and the curve was resampled
with points every 5 mm to represent regions along the length of the bundle. Then, this prototype
curve was transformed into subject native space using the transform computed using DTI-TK. For
a given tract, streamline vertices are reparameterized to best match the prototype to allow for
correspondence across streamlines at different cross-sections. Then, the average scalar parameter
was computed for each group of vertices that matched the vertices of the reference, and the
resulting along-bundle parameter maps were retained for subsequent statistical analysis.

Statistical analysis

For whole-tract analyses, males and females were combined because no significant
differences were identified (Supplemental Table 2.1). To assess gross patterns of white matter
tract development and provide overall tract summary parameters, the NDI/ODI values were
averaged over left and right sides for whole-tract analyses (e.g., Left ATR + Right ATR). In order
to better localize the core of a tract, tract density was used to derive per tract weighted averages of
NODDI parameters according to:

𝑥 ̅ =
∑ 𝑤 𝑖 𝑥 𝑖 𝑛 𝑖 =1
∑ 𝑤 𝑖 𝑛 𝑖 =1
(4)

where n denotes the number of voxels within a tract and wi and xi are the number of streamlines
passing through and the calculated NODDI parameter for the ith voxel within a tract, respectively,
and 𝑥 ̅
is the weighted parameter average for a tract.
The relationship between tract parameter weighted averages and age for each bilateral
whole-tract analysis were tested using linear, quadratic, logistic, Brody, von Bertalanffy, and
negative exponential growth curve models. Model selection using Bayesian information criteria

48

(BIC) was used to identify the curves that explained the most age-related variance in NODDI ODI
and NDI. Linear regression provided the best fit model for ODI and the Brody growth function
provided the best fit model for NDI and these were used in subsequent analyses. Brody growth
models of the form:

𝑁𝐷𝐼 = 𝛼 − (𝛼 − 𝛽 )𝑒 −𝑘 ∗𝑎𝑔𝑒 (5)

were fit to relate tract NDI and age using nonlinear least squares regression where α is the
asymptote, β is the y-intercept, and k is the growth rate.
Developmental timing was defined as the age at which NDI reaches 90% of its asymptotic
value, and the degree of maturation was calculated as the percent change in NDI at this terminal
age. Standard errors and confidence intervals for the coefficients, terminal maturation age, and
percent NDI change for each tract in equation were calculated using bootstrap resampling with
5000 iterations using R.
For along-tract analyses, Brody growth curves were fit at each point along the left and right
tracts separately to quantify the relationship between regional NDI and age. At each point, the age
at 90% maturation was calculated. Bootstrap resampling was performed along each tract with 5000
iterations to obtain a confidence interval around the regional along-tract age estimates.

2.4 Results

Mean ODI changes with age

There were weak but significant correlations between age and ODI in the CST, β=-.0013,
t(102)=-2.82, p=.005, forceps major, β=-.0034, t(102)=-3.31, p=.001, and IFO, β=-.0017, t(102)=-
2.54, p=.013, with ODI negatively associated with age in these regions. No other major white
matter tract showed age-related changes in mean tract ODI. When comparing across tracts, the
bilateral CST had the lowest average ODI (M=.258), while the bilateral CGC had the highest
average ODI (M=.387).

49

Table 2.1 Description of NODDI parameters in bilateral whole-tract analyses.
NDI ODI
Tract α (SE) t exp (SE)
Absolute
change
Percent
change T exp (SE) Mean (SE)
(0 - 18 years) (0 - 18 years)
ATR 0.580 (.034) 0.162 (.051) 0.271 92 9.85 (4.01) 0.335 (.002)
CGC 0.548 (.955) 0.131 (.088) 0.225 75 11.55 (194.36) 0.387 (.005)
CST 0.751 (.027) 0.201 (.053) 0.347 88 7.73 (2.11) 0.258 (.002)
FMAJ 0.619 (.182) 0.250 (.108) 0.287 87 6.17 (6.55) 0.264 (.005)
FMIN 0.530 (.012) 0.400 (.086) 0.328 162 4.55 (.97) 0.295 (.004)
IFO 0.546 (.022) 0.185 (.050) 0.258 93 8.60 (2.58) 0.298 (.003)
ILF 0.542 (.035) 0.156 (.052) 0.249 90 10.15 (5.12) 0.329 (.003)
SLF 0.622 (.032) 0.195 (.056) 0.317 108 8.51 (3.21) 0.339 (.003)
UNC 0.459 (.031) 0.176 (.077) 0.18 66 8.03 (5.08) 0.343 (.003)
Exponential fits relating age and NDI for asymptote (α) and exponential time constant (t exp) parameters,
absolute change of NDI per year and percent change per year from age .1 to 18.8 years, age where NDI
reaches 90% of the asymptotic value (T exp), and mean ODI averaged across whole tracts (left + right
combined). For all estimated parameters, the standard error is provided (SE).
ATR, anterior thalamic radiation; CGC, cingulum; CST, corticospinal tract; FMAJ, forceps major; FMIN,
forceps minor; IFO, inferior fronto-occipital fasciculus; ILF, inferior longitudinal fasciculus; SLF, superior
longitudinal fasciculus; UNC, uncinate fasciculus

Mean NDI Changes with Age

Significant exponential changes, as measured with Brody growth curves, were observed in
all bilateral tracts (Figure 2.2). The mean terminal ages estimated from the Brody model for each
tract ranged from childhood through adolescence (4.55-11.55 years, M=8.35 years). FMIN reached
terminal NDI maturation earliest (M=4.55 years, SD = .97 years) and the ILF reached terminal
NDI maturation latest (M=10.15 years, SD=5.1 years). The CGC was estimated to reach terminal
NDI latest (M=11.55), but bootstrapping yielded biologically implausible standard errors for the
upper asymptote (α) coefficient and age to reach 90% of α, (Table 2.1). The UNC showed the
smallest percent change in NDI from infancy through 18 years of age (66%), while the largest
percent change in NDI was observed in the FMIN (162%). Variance in terminal age estimates
varied across tracts, with the smallest variance observed in the FMIN (0.97) and the largest
variance observed in the FMAJ (6.55).

50

Figure 2.2 Age-related NDI increases in major white matter tracts with exponential fits.
NDI values are averaged across the left and right hemispheres for each tract. Maximum NDI
plateau of each tract are marked by the vertical dotted red lines and plots are ordered according to
how quickly the tract reaches the NDI plateau. CGC is not marked because it reaches the NDI
plateau outside the age range explored. There are differences between the tracts in the fit
parameters presented in Table 2.1.

Developmental Rate and Timing of Tract-Specific NDI and ODI Changes

The regional variability of ODI distribution and NDI developmental rates within individual
white matter tracts can be represented with a vector of measures sampled at equidistant points
along the length of each tract. Average ODI varied along the lengths of the tract and tended to be
higher in superficial tract segments compared to the core (Supplemental Figure 2.1). The ODI
distribution along each tract did not noticeably differ from the contralateral tract. For the tracts
explored, regional ODI was not significantly associated with age and could not be described using
a growth curve.

51

Figure 2.3 Along-tract changes in NDI and ODI within the forceps minor.
(A) Axial view of forceps minor tractography derived from the population template. The cord in
white is the prototypical streamline with maximum tract density and nodes represent discrete
sampling point locations along the tract. (B) The estimated age where NDI reaches 90% maturation
is plotted for each point along the length of FMIN from the left frontal cortex to the right frontal
cortex. Error bars indicate the bootstrapped 99% confidence interval. The developmental timing
of NDI is similar between the left and right sides of FMIN and the genu of the corpus callosum
reaches terminal NDI earliest compared to the rest of FMIN and is indicated with the yellow
asterisk in (A) and (B). (C) The average ODI is plotted for each point along the length of FMIN.
The shaded region indicates the standard deviation.

Average NDI varied along the lengths of each tract. In the majority of tracts, the
relationship between age and NDI at each point can be described with an exponential growth curve.
However, a multitude of regions within the left and right CGC did not show significant exponential
growth of NDI with age and were excluded from the following analysis. The estimated terminal
ages for FMAJ yielded unreliable age estimates well outside the age range for this study
(Supplemental Figure 2.2). Brody growth models applied to the resulting tracts demonstrated that
the estimated age at terminal NDI varies along the length of each tract. The estimated age to
maximum NDI tends to be higher and more variable in the superficial white matter compared to
the tract core. In general, regions that have younger terminal age estimates have smaller
bootstrapped variance than regions that have older terminal age estimates.
Some tracts have similar patterns of development between the left and right hemispheres.
The left and right fibers within the ATR and FMIN exhibit similar distributional patterns of
maturational timing (Supplemental Figure 2.2). ATR projections from the thalamus develop later
than the frontal cortical terminations bilaterally. Additionally, the developmental pattern of left
hemisphere FMIN fibers mirrors that of the right FMIN – NDI develops latest in the cortical
terminations to the frontal lobe, and gradually develops earlier as the fibers pass through the genu
of the corpus callosum, bilaterally (Figure 2.3). Other tracts showed lateralized patterns of NDI

52

development. In the right CST, the age at terminal maturation does not vary noticeably within the
tract core except for the cortical terminations. For the left CST, two regions develop later and with
more variability than the right – regions of the left CST that intersect with the centrum semiovale
and more inferiorally within the midbrain (Figure 2.4). The left and right SLF also shows
asymmetries in timing of NDI maturation. The superficial white matter of the left SLF is slightly
elevated compared to the deep white matter, which shows similar age estimates. The hemispheric
asymmetries observed in the CST and SLF cannot be explained by ODI variability, since regional
ODI is similar between the left and right hemispheres (Supplemental Figure 2.1, 2.2).

Figure 2.4 Along-tract changes in NDI and ODI within the corticospinal tract.
Tractography for the (A) left and (D) right corticospinal tracts derived from the population
template are shown in coronal slices. The cords in white are the prototypical streamline with
maximum tract density and nodes represent discrete sampling locations along the tract. The
estimated age where NDI reaches 90% maturation is plotted for each point along the (B) left and
(E) right corticospinal tract from the precentral gyrus descending inferiorly through the brainstem.
Error bars indicate the bootstrapped 99% confidence interval. The developmental timing of NDI
in the left (B) and right (E) corticospinal tract are not similar and there are two regions that develop
later in the left compared to the right side. These regions are indicated in (A) and (B) and include
fibers passing through the centrum semiovale (red asterisk) and at the level of the midbrain (orange
asterisks). The average ODI is plotted for each point along the (C) left and (F) right corticospinal
tracts and the shaded region indicates the standard deviation. The ODI distribution profiles are
similar between the left and right tracts.

53

2.5 Discussion

The current study sought to characterize the maturation patterns of major white matter
tracts in typical development from infancy through adolescence using NODDI. Using a semi-
automated tractography approach, we demonstrate widespread age-related increases in tract NDI.
NDI follows an exponential growth pattern and increases rapidly in infancy and young childhood
followed by a plateau in adolescence. NDI represents the tissue volume fraction occupied by
neurites, such as axons and dendrites, and is considered a proxy for neurite density (Sepehrband
et al. 2015). Postmortem evidence suggests that increases in neurite density, through myelination
and axon packing, is a prominent feature of white matter development in childhood (Benes 1989;
Miller et al. 2012b), and this is supported by findings that NDI accounts for more age-related
variance compared to DTI (Genc et al. 2017). Together, these results demonstrate NDI’s sensitivity
to developmentally relevant processes. ODI, a measure of axonal orientation variance, was not
significantly associated with age in the majority of white matter tracts, however weak negative
correlations were observed between ODI and age in the CST, FMAJ, and IFO. Our results are
largely in line with previous cross-sectional findings of positive correlations between NDI and age
and little or no correlation between ODI and age (Chang et al. 2015; Mah et al. 2017; Geeraert et
al. 2019), suggesting this period is marked by increased neurite density that are not accompanied
by changes in geometric complexity.
The magnitude and timing of neurite density maturation, as implied by cross-sectional
changes of NDI with age, can be obtained for each white matter tract using the growth curve fit
for that particular region. Significant variability was observed in the developmental trajectories of
whole-tract averages, with callosal fibers developing earlier than association tracts. The age where
maximum NDI is achieved can be defined as the time to reach 90% of asymptote estimated from
the individual tract growth curves. The FMIN and FMAJ, callosal connections critical for
interhemispheric communication (Aboitiz and Montiel 2003), show the earliest changes of NDI
with age, reaching terminal NDI before 7 years of age. In particular, FMIN, which connects
bilateral fronto-temporal regions through the genu of the corpus callosum, demonstrates the
earliest and most rapid NDI maturation, with NDI more than doubling by the time it reaches its
developmental plateau. These results are in agreement with previous findings demonstrating early
maturation of callosal projections using FA as a measure of microstructure (Lebel et al. 2008,

54

2019; Cancelliere et al. 2013; Chen et al. 2016; Krogsrud et al. 2016). NODDI studies have
demonstrated sparse growth (Chang et al. 2015; Geeraert et al. 2019) or small decreases (Mah et
al. 2017) of NDI with age in the FMIN, which may be attributed to the older age ranges used in
their samples (children older than 6 years of age).
The CST, a projection fiber critical for voluntary motor control, reached mature NDI next.
Association tracts developed last and include the fronto-temporal fibers of the UNC and SLF, the
fronto-occipital fibers of the IFO, the thalamo-cortical projections of the ATR, and the occipito-
temporal fibers of the ILF. The CGC, another important fronto-temporal connection that is part of
the limbic system, matures much later and reaches peak NDI after 18 years of age. These results
are consistent with a protracted pattern of frontal gray and white matter maturation observed in
several neurodevelopment diffusion studies (Lebel et al. 2008; Asato et al. 2010; Tamnes et al.
2010; Dean et al. 2015; Chen et al. 2016). Because there is evidence that regional developmental
patterns in white matter are linked to changes in behavior (Yeatman, Dougherty, Ben-Shachar, et
al. 2012; Klarborg et al. 2013; Peters et al. 2014; Ullman et al. 2014; Ullman and Klingberg 2016),
the variability in white matter development could provide important clues as to the concomitant
development of higher-level cognitive functions. These results are critical for understanding the
timing of white matter microstructural changes that shape emerging cognition.
While analyses of whole tract changes can inform the relative ordering of white matter tract
maturation, this approach may oversimplify white matter developmental processes by assuming
NODDI parameters change in unison across whole tracts. Along-tract analyses allow for the
quantification of tissue properties at discrete locations along the path of white matter fibers. Using
this approach, we demonstrate that aging does not affect major white matter tract NDI uniformly.
This is consistent with previous results showing regional patterns of white matter maturation using
diffusion parameters (Colby et al. 2011; Yeatman, Dougherty, Myall, et al. 2012; Johnson et al.
2014; Chen et al. 2016). In the present study, superficial white matter components of the tracts
tended to reach terminal NDI at older ages and with more variance compared to the core of the
tracts, as evidenced with the FMIN, left SLF, and right CST. The variability may be due in part to
partial volume effects at the interface of gray and white matter. However, as short, small-diameter
fibers that provide intra-cortical connections, superficial white matter has been shown to undergo
myelination later than deep white matter (Wu et al. 2014, 2016). ODI also varies along the length
of major white matter tracts, which can be explained by local differences in directional coherence,

55

including curvature, crossing fibers, and the entrance and exit of smaller axon bundles that
intersect with the tract. Different rates of NDI maturation may be explained by the
heterochronalogic development of white matter (Miller et al. 2012a), as the rate of myelination
can differ among white matter regions.
While the FMIN reached terminal NDI earliest, along-tract analyses reveal the fibers
passing through the genu of the corpus callosum develop earlier than the rest of the tract. While
previous studies on the development of diffusion parameters demonstrate early rapid growth
localized to the genu (Johnson et al. 2014; Chen et al. 2016), these findings could be attributed to
the coherent organization of callosal fibers that result in reduced diffusion signal contamination
(Genc et al. 2017). Our results suggest the early observed maturation of the genu using NDI, which
models the full spectrum of fiber orientations, is due to an increase in neurite density. The early
maturation of the corpus callosum relative to the rest of the brain may be due its role in
interhemispheric communication, which allows for functional integration, hemispheric
specialization, and information transfer between homologous cortical areas (Aboitiz and Montiel
2003), and there is some evidence supporting a general central-peripheral maturational gradient in
the brain (Yakovlev and Lecours 1967; Deoni et al. 2012).
Lateralization in the timing of white matter maturation was observed in the CST and this is
supported by previous longitudinal studies of whole-tract development reporting higher rates of
change of diffusion parameters in the left CST compared to the right (Lebel and Beaulieu 2011;
Krogsrud et al. 2016). In the present study, the developmental plateau for the right CST was
approximately uniform (5 years old), with the exception of the superficial white matter adjacent to
the precentral gyrus. In the left CST, we identified regions of fibers with prolonged and highly
variable developmental timing in locations near the centrum semiovale (20 years of age) and more
inferiorally at the level of the midbrain (16 years of age). Interestingly, these regions contain
crossing fibers and decussations and it is possible that the developmental variability observed may
be due to the inclusion of separate tracts intersecting with the CST. The major component of the
CST runs between motor regions in the frontal lobule and the brainstem before decussating and
innervating the contralateral side of the body. The delayed maturation of left CST subregions may
be attributed to the progressive refinement of fine and gross motor skills across development as
children acquire functional specialization of handedness preference.

56

Developmental asymmetries were also observed for the SLF. Whereas the rate of NDI
maturation was uniform across the core of the left SLF (9 years of age), the core of the right SLF
shows delayed maturation compared to the rest of the tract. The fronto-parietal fibers of the SLF
are implicated in diverse higher order cognitive processes, including language processing,
attention, and executive function (Catani 2007; Parker 2005; Urger et al, 2015). The protracted
maturation of NDI in the core of the right SLF may relate to functional lateralization of visuo-
motor processing (Budisavljevic et al. 2017), however additional evidence of SLF asymmetries
are mixed (Yin et al, 2013; Choi et al, 2010).
A main limitation of this study is the absence of participants older than 18 years of age,
potentially resulting in a selection bias that can influence the estimated age-related changes. The
exponential model used to characterize maturational trajectories may underestimate the age at the
developmental plateau because it is possible that the asymptote has not been achieved in the
majority of tracts by this time. Lifespan studies of white matter development show myelination
continues well into adulthood (Sowell et al. 2003; Hasan et al. 2010; Westlye et al. 2010; Lebel et
al. 2012) and diffusion studies demonstrate prolonged maturation of white matter FA (Lebel 2008;
Chen 2016), however it is unclear whether NDI maturation follows a shorter time course.
Another limitation of the present study was the cross-sectional nature of the design. Cross-
sectional studies can capture relationships between microstructural parameters and age, however
they are heavily influenced by inter-subject variability. Future longitudinal studies of white matter
maturation using NODDI models and along-tract analyses can provide the added benefit of
providing information about microstructural changes within subjects to more accurately define
white matter developmental trajectories.
The biophysical interpretation of NODDI parameters should be considered with caution
because, like all models, certain assumptions are made that may oversimplify tissue
microstructure. The minimal model employed by NODDI allows for reduced computation time,
but can result in bias and uncertainty with parameter estimation (Jelescu et al. 2016). For example,
NODDI fixes intra- and extra-axonal diffusivities to a priori values. This may bias the results
because some biological variability is expected that may change across developmental stages
(Jelescu et al. 2014, 2016; Kunz et al. 2014). However, NODDI provides some distinct advantages
over other multi-compartment methods, including its ability to model the full spectrum of neurite
orientations and ability to overcome partial voluming (Jelescu et al. 2014).

57

In conclusion, this study used NODDI parameters sensitized to microstructural features to
investigate the magnitude and timing of maturational changes in major white matter tracts from
infancy through adolescence. Along-tract analyses provide a framework to uncover spatial
distributions of microstructural age effects within white matter tracts, and our results indeed
demonstrate regional patterns of NDI timing within and across tracts. Additionally, ODI did not
show significant associations with age, suggesting that white matter development over this time
period is attributed to increases in neurite density, likely through increased axonal packing and
myelination. Overall, our results demonstrate the utility of NODDI models for characterizing the
heterochronous developmental patterns of white matter microstructure. These results add to the
growing body of research aiming to characterize how changes in brain structure contribute to
complex cognitive function, as these efforts are necessary for ultimately understanding the origins
of developmental disability.

2.6 References

Aboitiz F, Montiel J. 2003. One hundred million years of interhemispheric communication: The
history of the corpus callosum. Brazilian J Med Biol Res. 36:409–420.
Asato MR, Terwilliger R, Woo J, Luna B. 2010. White matter development in adolescence: A
DTI study. Cereb Cortex. 20:2122–2131.
Assaf Y, Cohen Y. 2000. Assignment of the water slow-diffusing component in the central
nervous system using q-space diffusion MRS: Implications for fiber tract imaging. Magn
Reson Med. 43:191–199.
Battaglini M, Smith SM, Brogi S, De Stefano N. 2008. Enhanced brain extraction improves the
accuracy of brain atrophy estimation. Neuroimage. 40:583–589.
Beaulieu C. 2002. The basis of anisotropic water diffusion in the nervous system - A technical
review. NMR Biomed. 15:435–455.
Beaulieu C. 2009. The biological basis of diffusion anisotropy. In: In: Johansen-Berg, H.,
Behrens TEJ (Eds. ., editor. Diffusion MRI: From Quantitative Measurement to In-vivo
Neuroanatomy. Academic Press. p. 105–129.
Behrens TEJ, Berg HJ, Jbabdi S, Rushworth MFS, Woolrich MW. 2007. Probabilistic diffusion
tractography with multiple fibre orientations: What can we gain? Neuroimage. 34:144–155.

58

Benes F. 1989. Myelination of cortical-hippocampal relays during late adolescence. Schizophr
Bull. 15:585–593.
Budisavljevic S, Acqua FD, Zanatto D, Begliomini C, Miotto D, Motta R, Castiello U. 2017.
Asymmetry and structure of the fronto-parietal networks underlie visuomotor processing in
humans. Cereb Cortex. 27:1532–1544.
Cabeen RP, Bastin ME, Laidlaw DH. 2016. Kernel regression estimation of fiber orientation
mixtures in diffusion MRI. Neuroimage. 127:158–172.
Cabeen RP, Laidlaw DH, Toga AW. 2018. Quantitative imaging toolkit: Software for interactive
3D visualization, data exploration, and computational analysis of neuroimaging datasets.
Cancelliere A, Mangano FT, Air EL, Jones B V., Altaye M, Rajagopal A, Holland SK, Hertzler
DA, Yuan W. 2013. DTI values in key white matter tracts from infancy through
adolescence. Am J Neuroradiol. 34:1443–1449.
Catani M, de Schotten MT. 2012. Atlas of human brain connections. Oxford University Press.
Catani M, Howard RJ, Pajevic S, Jones DK. 2002. Virtual in Vivo Interactive Dissection of
White Matter Fasciculi in the Human Brain. Neuroimage. 17:77–94.
Chang YS, Owen JP, Pojman NJ, Thieu T, Bukshpun P, Wakahiro MLJ, Berman JI, Roberts
TPL, Nagarajan SS, Sherr EH, Mukherjee P. 2015. White Matter Changes of Neurite
Density and Fiber Orientation Dispersion during Human Brain Maturation. PLoS One.
10:e0123656.
Chen Z, Zhang H, Yushkevich PA, Liu M, Beaulieu C. 2016. Maturation Along White Matter
Tracts in Human Brain Using a Diffusion Tensor Surface Model Tract-Specific Analysis.
Front Neuroanat. 10:1–18.
Clayden JD, Jentschke S, Muñoz M, Cooper JM, Chadwick MJ, Banks T, Clark C a., Vargha-
Khadem F. 2012. Normative development of white matter tracts: Similarities and
differences in relation to age, gender, and intelligence. Cereb Cortex. 22:1738–1747.
Colby JB, Soderberg L, Lebel C, Dinov ID, Thompson PM, Sowell ER. 2012. Along-tract
statistics allow for enhanced tractography analysis. Neuroimage. 59:3227–3242.
Colby JB, Van Horn JD, Sowell ER. 2011. Quantitative in vivo evidence for broad regional
gradients in the timing of white matter maturation during adolescence. Neuroimage. 54:25–
31.

59

Dean DC, O’Muircheartaigh J, Dirks H, Waskiewicz N, Walker L, Doernberg E, Piryatinsky I,
Deoni SCL. 2015. Characterizing longitudinal white matter development during early
childhood. Brain Struct Funct. 220:1921–1933.
Deoni SCL, Dean DC, Muircheartaigh JO, Dirks H, Jerskey BA. 2012. Investigating white
matter development in infancy and early childhood using myelin water faction and
relaxation time mapping. Neuroimage. 63:1038–1053.
Eluvathingal TJ, Hasan KM, Kramer L, Fletcher JM, Ewing-Cobbs L. 2007. Quantitative
diffusion tensor tractography of association and projection fibers in normally developing
children and adolescents. Cereb Cortex. 17:2760–2768.
Geeraert BL, Lebel RM, Lebel C. 2019. A multiparametric analysis of white matter maturation
during late childhood and adolescence. Hum Brain Mapp. [Epub ahea.
Genc S, Malpas CB, Holland SK, Beare R, Silk TJ. 2017. Neurite density index is sensitive to
age related differences in the developing brain. Neuroimage. 148:373–380.
Gogtay N, Giedd JN, Lusk L, Hayashi KM, Greenstein D, Vaituzis AC, Iii TFN, Herman DH,
Clasen LS, Toga AW, Rapoport JL, Thompson PM. 2004. Dynamic mapping of human
cortical development during childhood through early adulthood. PNAS. 101:8174–8179.
Goodlett CB, Fletcher PT, Gilmore JH, Gerig G. 2009. Group analysis of DTI fiber tract
statistics with application to neurodevelopment. Neuroimage. 45:S133–S142.
Hasan KM, Kamali A, Ewing-cobbs JMFL. 2010. Quantification of the spatiotemporal
microstructural organization of the human brain association , projection and commissural
pathways across the lifespan using diffusion tensor tractography. 361–373.
Holland S, Schmithorst VJ, Wagner M, Lee G, Rajagopal A, Sroka M, Felicelli N, Rupert A,
Clark K, Toga A, Freund L. 2015. The C-MIND project: normative MRI and behavioral
data from children from birth to 18 years. In: Proceedings of the 21st Annumal Meeting of
the Organization for Human Brain Mapping (OHBM), Honolulu, HA.
Huttenlocher PR. 1979. Synaptic density in human frontal cortex - Developmental changes and
effects of aging. Brain Res. 163:195–205.
Imperati D, Colcombe S, Kelly C, Di Martino A, Zhou J, Castellanos FX, Milham MP. 2011.
Differential Development of Human Brain White Matter Tracts. PLoS One. 6:e23437.

60

Jelescu IO, Veraart J, Adisetiyo V, Milla S, Novikov DS, Fieremans E. 2014. One diffusion
acquisition and different white matter models: How does microstructure change in human
early development based on WMTI and NODDI? Neuroimage. 107:242–256.
Jelescu IO, Veraart J, Fieremans E, Novikov DS. 2016. Degeneracy in model parameter
estimation for multi-compartmental diffusion in neuronal tissue. NMR Biomed. 29:33–47.
Jenkinson M, Bannister P, Brady M, Smith S. 2002. Improved optimization for the robust and
accurate linear registration and motion correction of brain images. Neuroimage. 17:825–
841.
Johnson RT, Yeatman JD, Wandell BA, Buonocore MH, Amaral DG, Nordahl CW. 2014.
Diffusion properties of major white matter tracts in young, typically developing children.
Neuroimage. 88:143–154.
Jones DK, Cercignani M. 2010. Twenty-five pitfalls in the analysis of diffusion MRI data. NMR
Biomed. 23:803–820.
Jones DK, Knösche TR, Turner R. 2013. White matter integrity, fiber count, and other fallacies:
the do’s and don’ts of diffusion MRI. Neuroimage. 73:239–254.
Kelly CE, Thompson DK, Chen J, Leemans A, Adamson CL, Inder TE, Cheong JLY, Doyle
LW, Anderson PJ. 2016. Axon density and axon orientation dispersion in children born
preterm. Hum Brain Mapp. 37:3080–3102.
Klarborg B, Skak Madsen K, Vestergaard M, Skimminge A, Jernigan TL, Baaré WFC. 2013.
Sustained attention is associated with right superior longitudinal fasciculus and superior
parietal white matter microstructure in children. Hum Brain Mapp. 34:3216–3232.
Kodiweera C, Alexander AL, Harezlak J, McAllister TW, Wu Y-C. 2015. Age effects and sex
differences in human brain white matter of young to middle-aged adults: A DTI, NODDI,
and q-space study. Neuroimage.
Krogsrud SK, Fjell AM, Tamnes CK, Grydeland H, Mork L, Due-Tønnessen P, Bjørnerud A,
Sampaio-Baptista C, Andersson J, Johansen-Berg H, Walhovd KB. 2016. Changes in white
matter microstructure in the developing brain-A longitudinal diffusion tensor imaging study
of children from 4 to 11years of age. Neuroimage. 124:473–486.
Kunz N, Zhang H, Vasung L, O’Brien KR, Assaf Y, Lazeyras F, Alexander DC, Hüppi PS.
2014. Assessing white matter microstructure of the newborn with multi-shell diffusion MRI
and biophysical compartment models. Neuroimage. 96:288–299.

61

Lebel C, Beaulieu C. 2011. Longitudinal development of human brain wiring continues from
childhood into adulthood. J Neurosci. 31:10937–10947.
Lebel C, Caverhill-Godkewitsch S, Beaulieu C. 2010. Age-related regional variations of the
corpus callosum identified by diffusion tensor tractography. Neuroimage. 52:20–31.
Lebel C, Gee M, Camicioli R, Wieler M, Martin W, Beaulieu C. 2012. Diffusion tensor imaging
of white matter tract evolution over the lifespan. Neuroimage. 60:340–352.
Lebel C, Treit S, Beaulieu C. 2019. A review of diffusion MRI of typical white matter
development from early childhood to young adulthood. NMR Biomed. 32:e3778.
Lebel C, Walker L, Leemans A, Phillips L, Beaulieu C. 2008. Microstructural maturation of the
human brain from childhood to adulthood. Neuroimage. 40:1044–1055.
Mah A, Geeraert B, Lebel C. 2017. Detailing neuroanatomical development in late childhood
and early adolescence using NODDI. PLoS One. 12:1–16.
Matsuzawa J, Matsui M, Konishi T, Noguchi K, Gur RC, Bilker W, Miyawaki T. 2001. Age-
related Volumetric Changes of Brain Gray and White Matter in Healthy Infants and
Children. Cereb Cortex. 11:335–342.
Miller DJ, Duka T, Stimpson CD, Schapiro SJ, Baze WB, McArthur MJ, Fobbs a. J, Sousa a.
MM, Sestan N, Wildman DE, Lipovich L, Kuzawa CW, Hof PR, Sherwood CC. 2012a.
Prolonged myelination in human neocortical evolution. Proc Natl Acad Sci. 109:16480–
16485.
Miller DJ, Duka T, Stimpson CD, Schapiro SJ, Baze WB, McArthur MJ, Fobbs AJ, Sousa
AMM, Sestan N, Wildman DE, Lipovich L, Kuzawa CW, Hof PR, Sherwood CC. 2012b.
Prolonged myelination in human neocortical evolution. Proc Natl Acad Sci U S A.
109:16480–16485.
Owen JP, Chang YS, Pojman NJ, Bukshpun P, Wakahiro MLJ, Marco EJ, Berman JI, Spiro JE,
Chung WK, Buckner RL, Roberts TPL, Nagarajan SS, Sherr EH, Mukherjee P, for the
Simons VIP Consortium. 2014. Aberrant white matter microstructure in children with
16p11.2 deletions. J Neurosci. 34:6214–6223.
Peters BD, Ikuta T, Derosse P, John M, Burdick KE, Gruner P, Prendergast DM, Szeszko PR,
Malhotra AK. 2014. Age-related differences in white matter tract microstructure are
associated with cognitive performance from childhood to adulthood. Biol Psychiatry.
75:248–256.

62

Pierpaoli C, Basser PJ. 1996. Toward a quantitative assessment of diffusion anisotropy. Magn
Reson Med. 36:893–906.
Rice D, Barone S. 2000. Critical periods of vulnerability for the developing nervous system:
Evidence from humans and animal models. Environ Health Perspect. 108:511–533.
Sepehrband F, Clark K a., Ullmann JFP, Kurniawan ND, Leanage G, Reutens DC, Yang Z.
2015. Brain tissue compartment density estimated using diffusion-weighted MRI yields
tissue parameters consistent with histology. Hum Brain Mapp. 00:n/a-n/a.
Sowell ER, Peterson BS, Thompson PM, Welcome SE, Henkenius AL, Toga AW. 2003.
Mapping cortical change across the human life span. Nat Neurosci. 6:309–315.
Stanisz GJ, Szafer A, Wright GA, Henkelman RM. 1997. An Analytical Model of Restricted
Diffusion in Bovine Optic-Nerve. Magn Reson Med. 37:103–111.
Tamnes CK, Ostby Y, Fjell AM, Westlye LT, Due-Tonnessen P, Walhovd KB. 2010. Brain
Maturation in Adolescence and Young Adulthood: Regional Age-Related Changes in
Cortical Thickness and White Matter Volume and Microstructure. Cereb Cortex. 20:534–
548.
Ullman H, Almeida R, Klingberg T. 2014. Structural Maturation and Brain Activity Predict
Future Working Memory Capacity during Childhood Development. J Neurosci. 34:1592–
1598.
Ullman H, Klingberg T. 2016. Timing of White Matter Development Determines Cognitive
Abilities at School Entry but Not in Late Adolescence. Cereb Cortex. 1–7.
Varentsova A, Zhang S, Arfanakis K. 2014. NeuroImage Development of a high angular
resolution diffusion imaging human brain template. Neuroimage. 91:177–186.
Veraart J, Sijbers J, Sunaert S, Leemans A, Jeurissen B. 2013. Weighted linear least squares
estimation of diffusion MRI parameters : Strengths , limitations , and pitfalls. Neuroimage.
81:335–346.
Wakana S, Caprihan A, Panzenboeck MM, Fallon JH, Perry M, Gollub RL, Hua K, Zhang J,
Jiang H, Dubey P, Blitz A, van Zijl P, Mori S. 2007. Reproducibility of quantitative
tractography methods applied to cerebral white matter. Neuroimage. 36:630–644.
Westlye LT, Walhovd KB, Dale AM, Bjørnerud A, Due-tønnessen P, Engvig A, Tamnes CK,
Østby Y. 2010. Life-Span Changes of the Human Brain White Matter : Diffusion Tensor
Imaging ( DTI ) and Volumetry. 2055–2068.

63

Wierenga LM, van den Heuvel MP, van Dijk S, Rijks Y, de Reus MA, Durston S. 2015. The
development of brain network architecture. Hum Brain Mapp. 37:717–729.
Woods RP, Grafton ST, Holmes CJ, Cherry SR, Mazziotta JC. 1998. Automated image
registration: I. General methods and intrasubject, intramodality validation. J Comput Assist
Tomogr. 22:139–152.
Wu M, Kumar A, Yang S. 2016. Development and aging of superficial white matter myelin from
young adulthood to old age: Mapping by vertex-based surface statistics (VBSS). Hum Brain
Mapp. 37:1759–1769.
Wu M, Lu LH, Lowes A, Yang S, Passarotti AM, Zhou XJ, Pavuluri MN. 2014. Development of
superficial white matter and its structural interplay with cortical gray matter in children and
adolescents. Hum Brain Mapp. 35:2806–2816.
Yakovlev P, Lecours A. 1967. The myelogenetic cycles of regional maturation of the brain.
Minkowski. ed, Regional development of the brain in early life. Oxford: Blackwell.
Yeatman JD, Dougherty RF, Ben-Shachar M, Wandell B a. 2012. Development of white matter
and reading skills. Proc Natl Acad Sci. 109:E3045–E3053.
Yeatman JD, Dougherty RF, Myall NJ, Wandell BA, Feldman HM. 2012. Tract Profiles of
White Matter Properties: Automating Fiber-Tract Quantification. PLoS One. 7.
Zhang H, Hubbard PL, Parker GJM, Alexander DC. 2011. Axon diameter mapping in the
presence of orientation dispersion with diffusion MRI. Neuroimage. 56:1301–1315.
Zhang H, Schneider T, Wheeler-Kingshott CA, Alexander DC. 2012. NODDI: practical in vivo
neurite orientation dispersion and density imaging of the human brain. Neuroimage.
61:1000–1016.
Zhang H, Yushkevich PA, Alexander DC, Gee JC. 2006. Deformable registration of diffusion
tensor MR images with explicit orientation optimization. Med Image Anal. 10:764–785.
Zhang S, Peng H, Dawe RJ, Arfanakis K. 2011. NeuroImage Enhanced ICBM diffusion tensor
template of the human brain. Neuroimage. 54:974–984.

64

2.7 Supplementary materials

Supplementary Table 3.1 Tests for significant laterality differences between left and right
whole-tract analyses and sex differences in left/right combined whole tract analyses
NOTE. n/a = not applicable
M=mean; SD=standard deviation
a
paired t-test (df=103)
b
two sample independent t-test (df=102)
*p<.05; **p<.01; ***p<.001

Left Right Females Males
Tract M SD M SD t
a
M SD M SD t
b

ATR 0.474 0.081 0.467 0.081 2.66** 0.472 0.073 0.477 0.086 -0.318
CGC 0.441 0.078 0.437 0.072 1.03 0.435 0.07 0.447 0.075 -0.84
CST 0.641 0.107 0.62 0.097 6.55*** 0.628 0.095 0.652 0.105 -1.217
FMIN n/a n/a n/a n/a n/a 0.479 0.078 0.475 0.09 0.241
FMAJ n/a n/a n/a n/a n/a 0.537 0.084 0.554 0.093 -0.971
IFO 0.468 0.078 0.448 0.071 8.5*** 0.452 0.072 0.461 0.077 -0.612
ILF 0.447 0.077 0.43 0.069 8.96*** 0.438 0.07 0.444 0.077 -0.414
SLF 0.53 0.099 0.496 0.09 10.68*** 0.51 0.086 0.525 0.099 -0.82
UNC 0.391 0.057 0.387 0.05 1.95 0.394 0.051 0.393 0.057 0.094

65

Supplementary Figure 1 Developmental timing of NDI maturation along white matter tracts.
The estimated age where NDI reaches 90% maturation is plotted for each point along the length
of bilateral major white matter tracts. Error bars indicate the bootstrapped 99% confidence interval.

66

Supplementary Figure 2 Mean ODI along the length of major white matter tracts.
The average ODI is plotted for each point along bilateral major white matter tracts. The shaded
region indicates the standard deviation.

67

Chapter 3
Hippocampal shape maturation in childhood and adolescence

Adapted from:
Lynch KL, Shi Y, Toga AW & Clark KA (2018), Cerebral Cortex

3.1 Abstract

The hippocampus is a subcortical structure critical for learning and memory, and a thorough
understanding of its neurodevelopment is important for studying these processes in health and
disease. However, few studies have quantified the typical developmental trajectory of the structure
in childhood and adolescence. This study examined the cross-sectional age-related changes and
sex differences in hippocampal shape in a multi-site, multi-study cohort of 1676 typically
developing children (ages 1-22 years) using a novel intrinsic brain mapping method based on
Laplace-Beltrami embedding of surfaces. Significant age-related expansion was observed
bilaterally and non-linear growth was observed primarily in the right head and tail of the
hippocampus. Sex differences were also observed bilaterally along the lateral and medial aspects
of the surface, with females exhibiting relatively larger surface expansion than males.
Additionally, the superior posterior lateral surface of the left hippocampus exhibited an age-sex
interaction with females expanding faster than males. Shape analysis provides enhanced sensitivity
to regional changes in hippocampal morphology over traditional volumetric approaches and allows
for the localization of developmental effects. Our results further support evidence that
hippocampal structures follow distinct maturational trajectories that may coincide with the
development of learning and memory skills during critical periods of development.

3.2 Introduction

The hippocampal formation is a subcortical brain structure within the medial temporal lobe
that plays a critical role in diverse learning and memory functions, including episodic memory
(Eldridge et al. 2005; Smith and Mizumori 2006), semantic learning (Henke et al. 1999; Manns et
al. 2003), and spatial navigation (Maguire et al. 1998; Burgess et al. 2002). The hippocampus also

68

supports broad cognitive processes, such as attention, perception, and social cognition (Tavares et
al. 2015; Aly and Turk-Browne 2017; Mack et al. 2017) and has been implicated in several
neurodevelopmental and psychiatric conditions, including schizophrenia (Pujol et al. 2014;
Haukvik et al. 2015), major depression (Bijanki et al. 2014; Maller et al. 2017), autism spectrum
disorders (Barnea-Goraly et al. 2014; Maier et al. 2015), and anxiety disorders (Machado-de-Sousa
et al. 2014; Lindgren et al. 2016). Because many of these disorders arise during childhood and
adolescence (Paus et al. 2008), the study of normative hippocampal development during this period
is important to understand atypical development.
The study of human hippocampal development in vivo is complicated by the heterogeneous
nature of the structure; the hippocampus is composed of several intricate subregions (Duvernoy et
al. 2013) and exhibits functional specialization along its long axis (Bannerman et al. 2004;
Poppenk et al. 2013; Strange et al. 2014). Hippocampal subregions include the subiculum, cornu
ammonis subfields (CA1, CA2, CA3, and CA4), and the dentate gyrus (DG) (Duvernoy et al.
2013). These subregions exhibit distinct cytoarchitectonic features and are differentially associated
with distinct aspects of memory formation (Reagh et al. 2014; Stokes et al. 2015; Dimsdale-Zucker
et al. 2018; Leal and Yassa 2018). Hippocampal subregions receive its major input from the
entorhinal cortex via the perforant pathway (Augustinack et al. 2010; Zeineh et al. 2016) and
information subsequently propagates through a series of connections within the hippocampus
(Parekh et al. 2015; Zeineh et al. 2016).
The topological and cellular features of the hippocampal subfields are established by birth
(Arnold and Trojanowski 1996; Insausti et al. 2010), however differential patterns of neuronal
proliferation, morphological maturation, and myelination occurs throughout childhood and
adolescence (Seress and Mrzljak 1992; Arnold and Trojanowski 1996; Seress 1998; Lavenex et
al. 2007; Lavenex and Banta Lavenex 2013; Dennis et al. 2016). These microstructural changes
may be reflected as alterations in regional hippocampal shape or volume during child development.
The functional gradient observed along the anterior-posterior long axis of the hippocampus is
reflected by differences in molecular expression (Sun et al. 2015), cell subtype distributions (Ding
and Van Hoesen 2015), gene expression (Fanselow and Dong 2010) and connectivity patterns
(Poppenk et al. 2013; Reagh and Ranganath 2018). The anterior hippocampus is preferentially
involved in anxiety-like behaviors and coarse, global memory representations, while the posterior
hippocampus is implicated in spatial learning and fine-grained, local representations (Bannerman

69

et al. 2004; Poppenk and Moscovitch 2011; Strange et al. 2014; Dandolo and Schwabe 2018;
Reagh and Ranganath 2018; Sekeres et al. 2018). Previous studies have shown that behavioral
performance in these domains are associated with structural and functional differences in the
anterior (Rajah et al. 2010; Zeidman et al. 2015) and posterior hippocampus (Maguire et al. 2000,
2006; Keller and Just 2016), which suggests that acquisition and refinement of these behaviors
during development may differentially alter hippocampal structure. Because memory function
improves rapidly from middle to late childhood (Ghetti and Angelini 2008; Ghetti and Lee 2011;
Ofen and Shing 2013), we expect structural hippocampal changes to co-occur during this period.
These maturational changes are commonly explored using whole hippocampal volumetry,
however previous studies of normative hippocampal volumetric development yield mixed results;
some studies have shown no bilateral age-related change in hippocampal volume (Yurgelun-Todd
et al. 2003; Gogtay et al. 2006; Knickmeyer and Gouttard 2008; Lin et al. 2013), while others have
found subtle increases in left, right, or bilateral hippocampal volume with age (Giedd et al. 1996;
Uematsu et al. 2012; DeMaster et al. 2014). It is likely that volumetric studies may obscure
developmentally relevant anatomical changes within specific subregions. Due to the structural and
functional heterogeneity observed along the longitudinal and cross-sectional axes, in vivo
structural MRI approaches measuring regional morphological changes may provide a more
complete understanding of hippocampal maturation.
It should be noted that some of the variability seen in previous studies may be due to advances
in and availability of MR hardware that influence scan quality. For example, the development and
widespread use of multi-channel radiofrequency receiver coils over the past decade have resulted
in the ability to acquire structural images with increased signal-to-noise ratio and higher spatial
resolution compared with single-channel coils (Wiggins et al. 2006), and, when combined with
parallel MRI acquisition techniques, significantly reduces acquisition time (Pruessmann et al.
1999; Ji et al. 2007). Now, because these technological advances have become commonplace,
researchers are afforded the tissue contrast and spatial resolution necessary in structural MR
images to identify subtle changes in hippocampal surface morphology across development.
In contrast to volumetric analyses, shape analysis techniques provide enhanced sensitivity to
regionally-specific surface deformations that may better reflect local structural changes in
development. The global geometry of elongated surfaces, such as the hippocampus, can be
quantified with radial distance measures, which maps the distance from a given vertex to the

70

medial core and is considered a regional index of “thickness” (Thompson et al. 2004; Shi et al.
2009). Global hippocampal shape features may also be a more informative biomarker than volume
for memory performance (Voineskos et al. 2015) and heritable traits (Sabuncu et al. 2016).
The first shape analysis studies of hippocampal development reported unique maturational
trajectories in individual subregions. Gogtay and colleagues (Gogtay et al. 2006) employed a
longitudinal design to explore regional hippocampal changes in a typically developing cohort
between 4 and 25 years of age and found differing maturational trajectories along the anterior-
posterior axis. While no changes in overall hippocampal volume with age were observed, shape
analysis revealed that posterior regions increased with age, while anterior regions decreased with
age. Another study, however, found age-related expansion of anterior hippocampal regions in a
more restricted age range between 6 and 10 years of age (Lin et al. 2013). These differences could
be attributed to different age ranges and small sample sizes. We propose to add to these important
findings by capitalizing on a large sample size to provide unprecedented power to detect regional
changes in hippocampal morphometry over a broad age range.
In this study, we sought to characterize regional hippocampal developmental trajectories and
sex differences in a large cross-sectional cohort of 1676 children and adolescents between 1 and
22 years of age using automated volumetric and shape analysis approaches. Localized age and sex
effects on intrinsic radial distance features of the hippocampal surface were explored using the
software Metric Optimization for Computational Anatomy (MOCA;
https://www.nitrc.org/projects/moca_2015/), which provides more accurate anatomical matching
because this approach has low edge distortion ratio errors compared with other surface analysis
techniques, including spherical demons and unit-sphere projections (Shi, Lai, et al. 2014). Through
this framework, we can explore regional hippocampal expansion and deformation in normative
development.

3.3 Materials and Methods

Subjects and MRI Acquisition

Structural T1-weighted MRI images and demographic data used in the preparation of this
article were obtained from 3 neuroimaging databases of typical child development and combined

71

into a single dataset to increase statistical power to detect developmental differences. These
databases include the Cincinnati MR Imaging of NeuroDevelopment (C-MIND) data repository,
the Philadelphia Neurodevelopmental Cohort (PNC) research initiative, and the Pediatric Imaging,
Neurocognition and Genetics (PING) Study database.

C-MIND dataset

Cross-sectional neuroimaging data from 80 typically developing participants were
considered from the C-MIND study. Structural MRI images were acquired with the same protocol
on a 3T Philips Achieva at Cincinnati Children’s Hospital (Holland et al. 2015). All MRI images
were manually checked for data quality and scans with excessive motion were discarded. Of the
80 subjects considered, 3 subjects were excluded due to poor data quality, resulting in 77 C-MIND
subjects for the present analysis (40 female, range: 1.4-18.8 years, M=9.0, SD=4.5 years).
Demographics of the subjects included is presented in Table 1.

PNC dataset

Structural neuroimaging data from 997 typically developing children and adolescents were
acquired by the PNC study at Children’s Hospital of Philadelphia (Satterthwaite et al. 2016). Scans
were acquired with the same protocol on a 3T Siemens Tim Trio whole-body MRI. Of the subjects
considered for the study, 46 subjects were excluded due to the quality control issues stated above,
resulting in 951 PNC subjects in the present analysis (450 female, range: 8.2-22.6 years, M=14.8,
SD=3.4 years) (Table 1).

PING dataset

At the time of collection, 801 structural MRI scans were available through the PING
database (http://ping.chd.ucsd.edu/) from a cohort of 695 typically developing children and
adolescents. Cross-sectional structural neuroimaging data were acquired from different scanner
vendors at 9 separate sites (Supplementary Table 3.1). PING was launched in 2009 by the

72

National Institute on Drug Abuse (NIDA) and the Eunice Kennedy Shriver National Institute of
Child Health & Human Development (NICHD) with the primary goal to create a data resource of
highly standardized and carefully curated MRI data. Of the 801 scans from the 695 subjects
considered for this study, 106 duplicate scans were removed and 47 scans were excluded due to
the quality control issues stated above, resulting in the inclusion of 648 cross-sectional scans (309
female, range: 3.2-22.6 years, M=11.9, SD=4.9 years) (Table 1).

Table 3.1 Study demographics by site
Study Sex n ICV (cm
3
) Age (years)
Mean SD t p Mean SD Range t p
CMN M 37 1108.4 100.7 10.14 4.48 2.5-17.5
F 40 1064.7 113.6 1.78 n.s. 7.93 4.36 1.42-18.83 2.19 *
PNC M 450 1195.3 104 14.43 3.35 8.25-21.75
F 501 1076.6 93.7 18.52 *** 15.06 3.44 8.33-22.58 2.85 **
PNG M 339 1188.5 110.9 11.84 4.86 3.17-21
F 309 1067.3 111.6 13.85 *** 11.96 5.04 3.17-21 0.31 n.s.
Combined M 826 1188.6 108.1 13.18 4.32 1.42-22.58
F 850 1072.6 101.5 22.63 *** 13.6 4.56 2.5-21.75 1.93 n.s.
Statistical tests compare ICV and age with sex (M, male; F, female). Positive t-statistics represent
M>F. C-MIND, Cincinnati MR Imaging of NeuroDevelopment dataset; PNC, Philadelphia
NeuroDevelopment Cohort; PING, Pediatric Imaging, Neurocognition and Genetics study; SD,
standard deviation; n.s., p>.05; *p<.05; **p<.01; ***p<.001

In total, 1676 children and adolescents between the ages of 1 and 22 years (mean age ±
standard deviation: 13.4 ± 4.5 years; 850 female) enrolled in C-MIND, PNC, or PING were
included in this study (Figure 3.1, Table 1). Throughout all protocols, participants had no self-
reported history of brain injury or major developmental, psychiatric, or neurological disorders.
Across all sites and studies, brain structural imaging was performed using three-dimensional
magnetization prepared, rapid-acquisition gradient-echo (MPRAGE) T1-weighted sequences on
3T MRI scanners with voxel sizes ranging from 0.9375 mm isotropic to 1x1x1.2 mm
3
. Scan
parameters for each site and study are provided in Supplementary Table 3.1. Because this study
used data from multiple scanners with slightly different spatial resolutions, all neuroimaging data
was first resampled to 0.9375 mm isotropic.

73

Of the 1772 cross-sectional MPRAGE scans considered from the 3 datasets, 29 scans were
removed for excessive head motion, 65 scans were removed for poor hippocampal segmentation,
and 2 scans were removed for anatomical abnormalities resulting in the 1676 scans mentioned
above.

Figure 3.1 Age distribution of participants included in the present study.
The histogram is color-coded according to study. CMN, Cincinnati MR Imaging of
NeuroDevelopment; PNC, Philadelphia Neurodevelopmental Cohort; PNG, Pediatric Imaging,
Neurocognition and Genetics study.

Image Processing
Across all datasets, image processing consisted of bias-field correction, registration to
standard MNI space, brain extraction and tissue-based segmentation using FSL’s anatomical
processing pipeline (Smith et al. 2004). Bilateral hippocampal volumes were identified in
structural MR images using FSL’s anatomical processing tool, FIRST (Patenaude et al. 2011).
FIRST is a fully-automated model-based subcortical segmentation approach that utilizes a
multivariate Gaussian framework to extract the most probable volumes from T1-weighted image
intensities.
These volumes were then binarized and total hippocampal volume was calculated. In order
to normalize for head size, intra-cranial volume (ICV) was computed per subject in cm
3
using the
number of voxels contained within the skull. Tissue segmentation in infants is made difficult due

74

to lower signal-to-noise ratio and poor tissue contrast compared with adult MR images (Shi et al,
2010, Shi et al, 2014). Consequently, T1-weighted images alone do not provide effective tissue
contrast in infants and segmentation is improved with the addition of T2-weighted images
(Williams et al, 2005; Guo et al, 2015). While T2-weighted images were not available in the
developmental databases used, steps were taken to ensure the quality of hippocampal
segmentations in the infant population analyzed. All T1-weighted images were manually checked
for data quality and scans with excessive motion were discarded. Additionally, all segmented
hippocampal volumes were visually inspected for anatomical accuracy and only well-delineated
structures were considered for analysis. Because tissue contrast in the brain improves after the first
year of life, analyses were replicated in a subset that excluded hippocampal segmentations from
infants less than 3.17 years of age (see below).

Hippocampal Mapping/Shape Analysis

Shape analysis was performed using the fully-automated Metric Optimization for
Computational Anatomy (MOCA) software developed by Shi and colleagues (Shi, Lai, et al. 2014)
(Figure 3.2). MOCA aligns anatomical features onto brain surfaces using Laplace-Beltrami (LB)
eigen-functions as isometry-invariant descriptors of surface geometry. Each hippocampal volume
was first converted to a triangulated mesh in native space by iteratively updating vertices through
outlier detection and surface deformation, resulting in robust topological preservation without
shrinkage (Shi, Lai, et al. 2010). In order to probe for inward and outward shape changes, intrinsic
local radial distance measures (RD) were determined using the Reeb graph of the first LB eigen-
function (Shi et al. 2008, 2009). This feature, which reflects

75

Figure 3.2 Shape Analysis method and hippocampal surface anatomy.
(A) Example hippocampal shape analysis procedure from 2 sample subjects. Hippocampal
volumes are segmented from the T1-weighted image using FSL. Each volume surface is converted
to a triangulated mesh in native space using MOCA and the radial distance is computed per vertex.
Individual meshes are then averaged together to generate a population-matched average atlas. A
pullback function is then applied to project individual radial distance metrics to the atlas surface
to allow for one-to-one correspondence of features. (B) The putative location of subfields are
shown on the average hippocampal surface using approximate geometric landmarks from
(Winterburn et al. 2013; Iglesias et al. 2015; Yushkevich, Amaral, et al. 2015). From left to right,
the right inferior posterior, right superior anterior, left superior anterior, and left inferior posterior
surfaces are shown.

76

local hippocampal thickness, was defined per subject as the shortest distance from each mesh
vertex to the long centroidal axis of the hippocampus.
All hippocampal meshes were averaged in a common space to form left- and right-
hemisphere hippocampal atlases with 2000 vertices per atlas using SurfStat implemented in Matlab
(www.math.mcgill.ca/keith/surfstat). The atlas mesh was then projected onto each subject surface
using conformal maps, resulting in one-to-one correspondence for statistical analyses.

Statistical Analysis

All analyses were performed separately on the left and right hippocampus. Because brain
size is significantly associated with age and sex during development, intracranial volume (ICV)
was used to calculate the adjusted hippocampal volume, Voladj (mm
3
), of each subject using the
following equation: Voladj = Volraw*(ICVm/ICVi), where Volraw is the raw hippocampal volume,
ICVm is the mean ICV across all subjects, and ICVi is the individual ICV. This approach allows
for hippocampal volumes to be expressed as a proportion of the occupied cranial cavity
(Voevodskaya et al. 2014; Nordenskjöld et al. 2015). In order to probe hippocampal anatomical
changes that co-occur with intra-cranial expansion, models without ICV correction were also
employed. General linear models were tested for the main effects of age and sex, and age*sex
interactions on hippocampal volume. Linear and quadratic age effects were tested on hippocampal
volume, and the best fit model was selected using Bayesian information criteria (BIC).
Similar to the volumetric adjustment described above, the per-vertex radial distance
measures obtained with shape analysis were adjusted for individual head size. For a given vertex
i, the adjusted radial distance, RDi,adj (mm), was calculated using: RDi,adj = RDi,raw*(ICVm/ICVi)
1/3
,
where RDi,raw is the raw radial distance for the ith vertex. Because RD reflects thickness (mm) as
opposed to volume (mm
3
), the ICV adjustment is scaled by the cube-root of total brain volume to
reflect radial thickness measures as a fraction of the total brain size (Kerchner et al. 2010;
Costafreda et al. 2011).
In order to determine regional variability in hippocampal development using shape
analysis, general linear models (GLMs) were applied to each adjusted and unadjusted vertex to
test for regional effects of age, sex, and age*sex interactions while controlling for scanner type
and site. Accordingly, age and sex were included as covariates in order to isolate the specific

77

effects of sex and age, respectively. Quadratic and cubic polynomial models were also explored to
identify non-linear age effects in hippocampal shape. To determine which GLM (linear, quadratic,
or cubic polynomial) explains the most age-related variance in radial distance measures, model
selection was performed for each significant vertex using BIC.
Multiple comparison techniques, such as Bonferroni correction, are not appropriate for
surface-based analyses because each vertex is not an independent observation – vertices are
spatially correlated with their neighbors. Random field theory (RFT) overcomes this by
considering both the peaks and spatial extent of smoothed statistical parameter maps using
Gaussian random fields (Cao and Worsley 1999; Worsley et al. 1999). The surface topology of
highly significant clusters is described by the Euler characteristic, which approximates the
corrected p-value at a given cluster level (Friston 1997; Woo et al. 2014). The expected Euler
characteristic is derived from the unadjusted p-value and the number of resels, or resolution
elements, in the image (Cao and Worsley 1999). This latter property depends on the smoothness
of the surface and number of observation points, and describes the search volume as a block of
vertices with the same size as the image smoothness FWHM (Worsley et al. 1992). A supra-
threshold cluster level of p<.001 and a set level threshold of p<.05 was specified to determine
height and spatial extent thresholds, respectively. Region of interest (ROI) post hoc analyses were
then performed on significant age-related clusters to identify the maturational trajectory of
developmentally relevant hippocampal regions.
Although all data was resampled to the same resolution, meshes obtained from automatic
hippocampal segmentation may be affected by differences in scanner contrast and partial volume
effects. To overcome this limitation, all volumetric and surface-based analyses used study and
scanner type as covariates, which has been shown to suppress the effects of scanner variability
(Fennema-Notestine et al. 2007; Pardoe et al. 2008; Chen et al. 2014; Takao et al. 2014). Because
all neuroimaging data from subjects younger than 3.17 years were acquired from the same study
and scanner (CMN), shape analysis was repeated on a subset of the dataset excluding these 10
infants and toddlers (N=1666) to determine whether this young population biased results (Figure
3.1). A marked improvement in T1-weighted tissue contrast is observed in participants greater than
1 year of age (Shi, Yap, et al. 2010; Guo et al. 2015) therefore analyses of this subset were
additionally motivated by the effect of tissue contrast on hippocampal segmentation. Analyses
were also performed with database-matched age ranges by removing scans acquired from

78

individuals less than 8.25 years of age since the PNC study does not include subjects under this
age (N=1469).

3.4 Results

Whole-Hippocampus Volumetric Results

Bilateral unadjusted hippocampal volumes exhibited significant non-linear increases with
age as represented with quadratic polynomials. These increases occurred rapidly in early life and
plateaued during adolescence. ICV-adjusted hippocampal volume was also significantly
associated with age, left: b=.019, t(1675)=8.02, p<.001, adjusted R
2
=0.04; right: b=.019,
t(1675)=8.06, p<.001, adjusted R
2
=0.04, exhibiting linear volumetric expansion during

Figure 3.3 Age-related changes in adjusted hippocampal volume stratified by sex.
Bilateral hippocampi adjusted for ICV exhibited sex differences and significant linear volumetric
expansion with age when controlling for scanner type and study. Hippocampal volume is
represented in cm
3
. Individual data points and regression lines are coded according to sex (blue,
males; red, females).

development with respect to head size (Figure 3.3). There was a large effect of hemisphere on
adjusted hippocampal volume, t(1676)=10.59, p<.001, d=1.9, with larger volumes in the right

79

hippocampus (M=3.65, SD=.43) compared to the left (M=3.56, SD=.44), and this relationship was
maintained for the unadjusted hippocampal volumes.

Table 3.2 Significant clusters related to age and sex using shape analysis
Hemisphere Test (x,y,z) Coordinates resels p-value Direction
Left Age (66.7, 126.9, 58.4) 32.62 <.001 Linear+
(66.0, 125.1, 55.1) 2.12 <.001 Quadratic+
(77.8, 120.6, 62.3) 1.23 <.001 Quadratic+
(65.8, 101.3, 66.9) 0.82 0.005 Quadratic+
(69.4, 111.1, 67.4) 0.42 0.03 Quadratic+
Sex (72.1, 106.2, 69.1) 3.64 <.001 F>M
(63.3, 111.1, 61.4) 7.15 <.001 F>M
Age*Sex (63.3, 103.1, 72.3) 0.94 0.002 Frate > Mrate
(71.6, 122.6, 55.4) 0.41 0.03 Frate > Mrate
Right Age (116.4, 97.4, 70.7) 32.77 <.001 Linear+
(108.7, 115.7, 61.2) 1.29 <.001 Linear+
(111.6, 103.8, 71.8) 12.5 <.001 Quadratic+
(112.5, 126.5, 53.7) 3.16 <.001 Quadratic+
(104.9, 126.2, 58.2) 0.83 0.005 Quadratic-
Sex (120.3, 114.7, 68.1) 10.68 <.001 F>M
(109.9, 106.3, 68.9) 6.53 <.001 F>M
(116.4, 126.5, 59.5) 1.23 <.001 F>M
Coordinates (x,y,z) correspond to the spatial location of the center of the cluster in Cartesian
coordinates on the group-averaged template; Resels, resolution elements; +, surface expansion; -,
surface contraction; F, female; M, male; Frate and Mrate are the slopes obtained by regressing local
surface changes on age for females and males, respectively. Adjusted cluster-wise p-values from
random field theory are provided

A medium effect of sex was observed in the unadjusted hippocampal volumes, left:
t(1675)=8.41, p<.001, d=.52, and right, t(1675)=8.53, p<.001, d=.51, with males (left: M=3.66,
SD=.46; right: M=3.75, SD=.47) having significantly larger hippocampi than females (left:
M=3.43, SD=.42; right: M=3.52, SD=.43). The nature of this relationship is reversed when
adjusting hippocampal volume for ICV (Figure 3.3), left: t(1675)=6.26, p<.001, d=.29, and right,

80

t(1675)=6.73, p<.001, d=.33, with females exhibiting small, but significantly larger proportional
hippocampal volumes (left: M=3.62, SD=.44; right: M=3.72, SD=.43) than males (left: M=3.49,
SD=.43; right: M=3.58, SD=.42). Neither adjusted nor unadjusted hippocampal volumes showed
age-by-sex interactions.

Age-Related Changes in Hippocampal Shape

Shape analysis results are presented in Table 3.2. Significant linear and nonlinear age-
related expansion (i.e. larger thickness) was observed bilaterally in the unadjusted hippocampal
surfaces (not shown). ICV adjustment did not change the maturational pattern of expansion,
therefore all results presented here utilized the adjusted hippocampal thickness measures unless
otherwise specified. Linear age-related expansion was observed primarily on the superior and
inferior surfaces of the hippocampal tail and posterior poles bilaterally. Linear expansion was also
observed on the lateral anterior surfaces, bilaterally, and the mesial anterior surface of the left
hippocampus (Figure 3.4; Table 3.2)
Quadratic main effect terms in age were also statistically significant in bilateral hippocampal
regions, though the effects were more widespread on the right hippocampus where significant
quadratic expansion was observed on the lateral surfaces of the hippocampal head and tail (Figure
3.4). BIC model selection further supports the use of quadratic age, instead of linear age, when
describing developmental shape changes in the right hippocampal tail and head (Figure 3.4).
While males and females both show statistically significant quadratic age expansion of the right
lateral hippocampal head, only females show quadratic age expansion in the right lateral
hippocampal tail (Table 3.3). While no linear age-related decrease in hippocampal thickness was
observed bilaterally, a U-shaped quadratic relationship with age was observed in the right medial
hippocampal head (Table 3.2). When stratified by sex, only females show significant inward
deformation of the right medial hippocampal head, suggesting that this effect is driven by females
(Table 3.3). Exclusion of participants younger than 3.17 years of age did not significantly change
the results (Supplementary Figure 3.1), however, exclusion of participants younger than 8.25
years of age showed primarily linear age-related expansion in the hippocampal tail and quadratic
age effects were not observed.

81

Figure 3.4 Linear and non-linear hippocampal expansion with age.
(A) P-value maps of linear and quadratic regression of age-related adjusted hippocampal surface
expansion using BIC model selection. From left to right, the right inferior posterior, right superior
anterior, left superior anterior, and left inferior posterior surfaces are shown. Blue clusters signify
linear expansion with age while red clusters show quadratic expansion with age. Quadratic
expansion accounted for more variance than linear expansion primarily on the right hippocampus.
All analyses controlled for sex, scanner type, and study. (B) From left to right, region-of-interest
analyses for significant age-related clusters on the right hippocampus show quadratic expansion
along the lateral head, y=5.63+4.88x-2.85x
2
, p<.001, adjusted R
2
=.060, linear expansion along the
body, y=3.79+.019x, p<.001, adjusted R
2
=.113, and quadratic expansion along the tail,
y=3.94+4.22x-1.75x
2
, p<.001, adjusted R
2
=.062.

Hippocampal Shape Sex Differences

Surface-based shape analysis revealed significantly larger surfaces in males along the majority of
the left and right unadjusted hippocampal surfaces compared to females, left cluster: 35.2 resels,
t(1674)=12.39, p<.001, d=.61; right cluster: 32.8 resels, t(1674)=10.52, p<.001, d=.51. When

82

adjusting for ICV, the opposite trend was found, though the effect size was smaller (Figure 3.5a).
Females exhibited significantly larger surfaces compared to males bilaterally along the
Table 3.3 Significant age-related clusters in males and females
Sex Hemisphere
(x,y,z)
Coordinates resels
p-
value Direction
Male Left (71.2, 98.3, 76.5) 11.58 <.001 Linear+
(65.8, 126.3, 58.7) 2.13 <.001 Linear+
(78.3, 122.4, 60.1) 1.58 <.001 Linear+
(65.4, 125.2, 55.5) 1.34 <.001 Quadratic+
Right (116.5, 98.9, 69.4) 6.71 <.001 Linear+
(113.3, 101.7, 73.6) 6.57 <.001 Linear+
(114.4, 128.2, 57.6) 2.5 <.001 Linear+
(113.7, 128.0, 58.2) 1.22 <.001 Quadratic+
Females Left (66.7, 126.9, 58.4) 29.06 <.001 Linear+
(65.7, 124.3, 55.2) 0.45 0.02 Quadratic+
Right (116.9, 96.8, 71.2) 30.77 <.001 Linear+
(118.5, 97.1, 75.1) 5.14 <.001 Quadratic+
(114.2, 125.7, 53.7) 2.56 <.001 Quadratic+
(109.1, 121.4, 64.8) 0.71 0.008 Quadratic+
(104.2, 127.1, 59.1) 0.61 0.01 Quadratic-
Coordinates (x,y,z) correspond to the spatial location of the center of the cluster in Cartesian
coordinates on the group-averaged template; Resels, resolution elements; +, surface expansion; -,
surface contraction; Adjusted cluster-wise p-values from random field theory are provided.

lateral and medial surface of the adjusted hippocampal body (Figure 3.5a), accounting for 30%
and 37% of the left and right hippocampal surfaces, respectively (Table 3.2). Neither the left nor
right adjusted surfaces displayed significantly larger surfaces in males compared to females.
Exclusion of participants younger than 3.17 and 8.25 years of age did not significantly change the
results (Supplementary Figure 3.2).
Age-Sex Interaction in Hippocampal Shape

83

A significant interaction between age and sex was observed in the superior posterior lateral
body of the left adjusted hippocampus. A region-of-interest analysis of the cluster shows that
females exhibit a greater rate of change in surface expansion compared to males (Figure 3.5b). In
early childhood, the superior posterior lateral body of the left adjusted hippocampus is larger in
males than females; however females surpass males in adjusted size during approximate
adolescence. When stratified by sex, this region does not significantly expand with age in males,
but does in females. No significant age-sex interactions were observed on the right adjusted

Figure 3.5 Main effect of sex and age*sex interaction on hippocampal shape.
(A) Significant F>M clusters and peaks for the main effect of sex extracted from the t-map using
a corrected p<.05 threshold when controlling for age, scanner type, and study. On the left surface,
a cluster on the lateral body, t(1674)=8.20, p<.001, d=.4, and mesial body, t(1674)=6.12, p<.001,
d=.3, reached statistical significance. On the right surface, clusters located on the lateral edge,
t(1674)=8.68, p<.001, d=.42, medial edge, t(1674)=6.76, p<.001, d=.33, and head, t(1674)=5.04,
p<.001, d=.25, reached significance. (B) Significant age*sex interaction on adjusted hippocampal
volume is observed on the superior posterior lateral body of the left hippocampal surface when
controlling for scanner type and study. Region-of-interest analysis of the significant cluster shows
the average adjusted radial distance regressed against age and stratified by sex. Females show a
significant and positive relationship with age, b=.018, t(847)=5.97, p<.001, while males do not,
b=.0002, t(824)=.37, p=.71. A corrected p<.05 threshold was applied to obtain the p-value surface
maps.

hippocampus or bilateral unadjusted surfaces. Exclusion of participants younger than 3.17 and
8.25 years of age did not significantly change the results (Supplementary Figure 3.3).

84

The timing and magnitude of left hippocampal shape changes at different developmental
time points in males and females are illustrated in Figure 3.6. Subjects were stratified by sex and
binned into age groups (3-5 years, 9-10 years, 14-15 years, and 19-20 years) where the average
difference in adjusted hippocampal thickness from baseline (3-5 years) was calculated. In females,
left hippocampal head and tail enlargement is apparent by 9-10 years of age, while expansion of
the left superior posterior lateral body becomes more apparent in adolescence (14-15 years). Males
exhibit modest left hippocampal expansion with age, with growth largely restricted to the head and
tail, though to a lesser extent than females.

3.5 Discussion

This is the largest study on hippocampal shape analysis in typical development to date,
with data from 1676 children and adolescents between the ages of 1 and 22 years of age, allowing
for unprecedented power to detect age- and sex-related differences in regional hippocampal
thickness. Bilateral hippocampal volumes increase linearly with age, however hippocampal shape
development is heterogeneous and dynamic. Hippocampal subregions follow distinct maturational
trajectories with posterior and anterior regions showing nonlinear changes over time primarily in
the right hippocampus and the lateral body showing linear expansion bilaterally. Additionally,
females showed relatively larger lateral and mesial subregions and more rapid maturation in left
posterior regions compared to males.
Although the source of volumetric expansion or contraction is unknown, these changes
may reflect differential expression of processes critical for postnatal hippocampal maturation, such
as neuronal proliferation, dendritic and axonal elaboration, synaptogenesis, or myelination (Seress
and Ribak 1995; Altemus et al. 2005; Lavenex et al. 2007; Insausti et al. 2010; Seress and Ábrahám
2014). Hippocampal structural maturation may also be influenced by neurogenesis, which occurs
within the subgranular zone (SGZ) of the postnatal mammalian hippocampus (Kornack and Rakic
1999; Seri et al. 2001; Spalding et al. 2013). While there is evidence that neurogenesis in the SGZ
is associated with certain structural and functional features in humans (Hueston et al. 2017; Powell
et al. 2017; Toda et al. 2018), recent studies have shown that primate neurogenesis declines rapidly
in early childhood (Spalding et al. 2013; Dennis et al. 2016) to undetectable levels by adulthood
(Sorrells et al. 2018). Hippocampal expansion during development may also provide more surface

85

area for maturing projections to and from cortical areas to support the acquisition and refinement
of hippocampal-dependent behaviors. This relationship between function and shape is supported
by evidence showing that hippocampal shape may be a more informative biomarker of working
memory performance than subfield volume (Voineskos et al. 2015).

Figure 3.6 Maturational trajectory of left hippocampal surface in males and females.
Data was stratified by sex and split into the following groups: 3-5 years (baseline), 9-10 years, 14-
15 years, and 19-20 years. For each age group and sex, the inferior posterior surface is presented
on the left side and the superior anterior surface is presented on the right side. Adjusted
hippocampal surfaces within each age range were averaged together from which the baseline was
subtracted to visualize the timing and magnitude of age-related hippocampal surface expansion.
Baseline (left-most surface) is illustrated in green. Surface expansion with respect to baseline is
mapped in warm colors while surface contract is mapped in cool colors.

In the present study, age was positively and linearly associated with thickness in superior
and inferior lateral posterior hippocampal regions, and these effects persisted when removing
subjects less than 8.25 years of age. Nonlinear age-related expansion was observed largely along
the right lateral posterior surface. When removing children younger than 8.25 years of age, only
the linear age effects in the posterior hippocampus were observed. This suggests developmental
expansion of the posterior hippocampus persists through late childhood and adolescence, which is

86

consistent with rapid improvements in memory performance during this period (Ghetti and
Angelini 2008). The developmental expansion in this region appears to correspond to putative
CA1 and subiculum subfields. The posterior hippocampus is selectively involved in spatial
learning and navigation (Maguire et al. 2000; Keller and Just 2016), perceptual learning (Sheldon
and Levine 2016), post-encoding processing (Poppenk and Moscovitch 2011), and the
representation of fine grained, local features (Reagh and Ranganath 2018; Sekeres et al. 2018).
These perceptual systems depend upon reciprocal connectivity between the posterior two-thirds of
the hippocampus and a broader posterior medial network, which includes the cuneus and
precuneus, posterior cingulate cortex, parahippocampal cortex, inferior parietal cortex, and
retrosplenial cortex (Kahn et al. 2008; Poppenk and Moscovitch 2011; Libby et al. 2012; Reagh
and Ranganath 2018; Sekeres et al. 2018). Maturation of these projections may therefore influence
the protracted structural development of the posterior hippocampus as reflected by increased
thickness and surface area.
The anterior hippocampus exhibited linear age-related expansion of the superior lateral
surfaces and nonlinear growth observed on the lateral edges of the hippocampal surface.
Interestingly, non-linear contraction of the surface was observed on the right mesial surface. These
regions likely correspond to CA1 and the subiculum, as the subiculum accounts for the largest
single cytoarchitectonic field in the hippocampal head (Insausti et al. 2010). The hippocampal head
exhibits a different connectivity profile compared to the posterior hippocampus (Poppenk et al,
2013). The anterior hippocampus is functionally correlated with a broader anterior temporal
network involved in conceptual processing, that includes the amygdala, medial prefrontal cortex,
lateral orbitofrontal cortex, and anterior temporopolar cortex (Kahn et al. 2008; Catenoix et al.
2011; Poppenk and Moscovitch 2011; Reagh and Ranganath 2018; Sekeres et al. 2018). In line
with this observed connectivity profile, the anterior hippocampus is preferentially involved in
anxiety-like behaviors (Satpute et al. 2012; Pantazatos et al. 2014), semantic encoding (Greve et
al. 2011), associative learning (Chua et al. 2007; Sheldon and Levine 2016), and processing coarse-
grained, global features during memory transformation (Sekeres et al. 2018). The heterogeneous
developmental trajectories in the right anterior hippocampus observed in the present study may be
attributed to the diverse functions carried out within the hippocampal head. Additionally, evidence
suggests that the functional organization of the hippocampus changes with time (DeMaster et al.
2014; Dandolo and Schwabe 2018). These findings may clarify previous hippocampal shape

87

analysis studies that have found either hippocampal head expansion (Lin et al. 2013) or contraction
(Gogtay et al. 2006) during development.
Another explanation for the simultaneous lateral expansion and mesial contraction of the
right hippocampal head could be that the anatomical location of the structure physically shifts
during development. Superior and medial to the hippocampal head lies the amygdala (Amunts et
al. 2005), which exhibits age-related volumetric expansion when controlling for total brain volume
(Østby et al. 2009) and right-greater-than-left structural laterality (Giedd et al. 1996; Uematsu et
al. 2012). Therefore, it is possible that the shift in right hippocampal structure occurs via lateral
displacement of the hippocampal head by the amygdala during development as they compete for
subcortical space. This is supported by evidence that brain structures undergo region-specific
maturation resulting in adult phenotypes that differ slightly in stereotactic location compared to
those observed in infants and children (Wilke et al. 2002, 2003; Yoon et al. 2009; Hill et al. 2010).
Overall, we observed that females have larger bilateral hippocampal volumes than males
when hippocampal volumes are adjusted for ICV. The present shape analysis results reveal that
the observed volumetric differences are localized to the lateral and mesial surfaces of the
hippocampal body. Females also exhibited age-related expansion within the superior posterior
body of the left hippocampus, while males showed no change in shape (Figure 3.5, Figure 3.6).
Previous studies have also reported sex differences in lateral and posterior hippocampal shape in
post-pubertal (Neufang et al. 2009; Satterthwaite et al. 2014), but not pre-pubertal (Satterthwaite
et al. 2014) maturation, suggesting that these sex differences do not arise until later in development.
Sex differences in hippocampal shape may be attributed to specific cellular mechanisms induced
by gonadal hormones (Bramen et al. 2011; Uematsu et al. 2012). Estrogen and androgen epitopes
are located throughout the hippocampus, including synapses, dendrites, terminals and glial cell
processes (McEwen and Milner 2007). In animal models, estrogen promotes synaptic plasticity in
hippocampal CA1 (Cooke and Woolley 2005; Yuen et al. 2011) and CA3 (Scharfman and
MacLusky 2017) pyramidal cells. It has also been shown that estrogen induces neurogenesis within
the hippocampal DG subgranular zone (Bowers et al. 2010) and may play a neuroprotective role
(Wise 2006; Bean et al. 2014). The relatively larger surfaces observed in females compared to
males in the present study may be due to estrogen-induced synaptogenesis and neural proliferation
within the hippocampus. However, the datasets used, with the exception of C-MIND, did not
include information on pubertal status or circulating hormone levels, which would have allowed

88

for increased specificity in the identification of the timing and magnitude of sex differences.
Therefore, conclusions regarding sex differences related to puberty await further investigation.
The results presented in this current study provide a basis for understanding normative
hippocampal development from childhood through adolescence. Our results support the
hypothesis that functionally distinct regions within the hippocampus exhibit different
developmental trajectories. These maturational changes may co-occur with acquisition of
hippocampal-dependent behaviors, however this question was not addressed in the current study.
Understanding normative hippocampal development is also important because it provides the basis
for identifying deviations from the expected typical developmental trajectory that may be
indicative of neurodevelopmental disability. Knowledge of the timing and magnitude of regional
hippocampal shape development may help inform the neurobiological underpinnings of disorders
associated with hippocampal pathology that arise during childhood and adolescence, including
schizophrenia (Paul and Harrison 2004), major depression (Bremner et al. 2000), autism spectrum
disorders (Saitoh et al. 2001; Philip et al. 2012), and substance use disorders (Berman et al. 2008).
For example, evidence shows inward deformation of the lateral anterior hippocampal head in
patients with childhood-onset schizophrenia compared to controls (Johnson et al. 2013), which
may map to anterior CA1 subregions (Kalmady et al. 2017). Results from the present study show
that this same region exhibits age-related expansion during normative development (Figure 3.4),
which suggests that patients with childhood-onset schizophrenia may have maturational
trajectories that differ from typically developing populations within the anterior head of the
hippocampus. Identification of the origin of this deviation in shape may aid in early diagnosis and
improved clinical outcomes in patients.
While the present study is the largest study of regional hippocampal development to date
and affords power to detect developmental changes and sex differences in a structure as variable
and heterogeneous as the hippocampus, some limitations exist. The study utilized a cross-sectional
design; therefore inferences on individual developmental trajectories cannot be realized without
providing longitudinal data points. The scans used in this study were obtained as a part of 3
separate datasets and acquired on different scanner platforms with slightly different acquisition
parameters (Supplementary Table 3.1), which may have influenced scan quality. While previous
studies have shown that differences in scanner version or manufacturer significantly influence gray
and white matter contrast ratios in T1-weighted images (Shuter et al. 2008; Shokouhi et al. 2011),

89

the estimated morphometric parameters, such as subcortical volume, do not change significantly
across platforms and field strengths in multi-site studies (Briellmann et al. 2001; Stonnington et
al. 2008; Jovicich et al. 2009, 2013; Segall et al. 2009). Additionally, study and scanner type were
used as covariates, which has been shown to suppress the effects of scanner variability (Fennema-
Notestine et al. 2007; Pardoe et al. 2008; Chen et al. 2014; Takao et al. 2014). Therefore, we do
not suspect that our segmentation process and the resulting shape analysis were compromised by
scanner differences.
The age distribution of the sample is highly skewed due to the prevalence of available scans
from participants greater than 8 years of age across all databases. We chose to include data from
all available ages in order to increase the sensitivity to detect developmental differences; therefore,
our results must be interpreted within the constraints of this limitation. Nevertheless, we replicated
analyses in subjects greater than 8.25 years of age and found retention of several clusters along the
tail that exhibit linear, outward growth over time.
While T1-weighted images provide good subcortical contrast in older children, delineation
of the infant hippocampus using structural MRI is challenging because the properties of the
developing brain, such as partial myelination, create poor tissue contrast (Shi, Yap, et al. 2010;
Shi, Wang, et al. 2014). A multi-modal approach to infant hippocampal segmentation utilizing T1-
and T2-weighted images would have improved shape analysis results due to the complementary
tissue information provided by these contrasts (Williams et al. 2005; Guo et al. 2015), however
T2-weighted images were not available from the databases used. To overcome this limitation, all
hippocampal segmentations were manually checked for segmentation accuracy and analyses were
performed excluding the youngest participants under 3 years of age and discovered the same
developmental pattern across the hippocampal surface. Because hippocampal structure changes
drastically within the first year of life (Insausti et al. 2010), future studies should seek to understand
how hippocampal shape changes during infancy utilizing methods more precisely tailored to the
unique challenges that come with segmenting developing neonatal tissue (Shi, Fan, et al. 2010;
Shi, Yap, et al. 2010; Shi, Wang, et al. 2014).
Shape analysis is sensitive to structural changes, however it does not provide specific
information regarding the nature of these changes. Because shape analysis techniques only
consider intrinsic surface features, differences in shape may reflect a number of structural
phenomena within the sampled space. Therefore, significant developmental differences at a given

90

vertex cannot be uniquely attributed to a specific subfield. The development of new hippocampal
segmentation techniques for structural MRI allows for subfield specificity (Winterburn et al. 2013;
Iglesias et al. 2015; Yushkevich, Pluta, et al. 2015); however these approaches may not provide
sufficient accuracy due to technical limitations on the spatial resolution required to resolve
individual subfield boundaries in clinically feasible scans (Amunts et al. 2005; Wisse et al. 2012;
Winterburn et al. 2013). Recent technological advances, including the increased availability of
ultra-high field MRI (Giuliano et al. 2017) and parallel imaging techniques utilizing multi-channel
receiver coils, such as simultaneous multi-slice imaging techniques (Barth et al. 2016; Setsompop
et al. 2016), provide sub-millimeter resolution at reduced acquisition times to enable detailed
subfield morphology (Giuliano et al. 2017). Future studies should aim to characterize how changes
in individual subfield volumes and morphometry influence regional features on the surface of the
hippocampus in order to better understand the underlying neurobiological changes in hippocampal
shape that occur during child and adolescent development.
The current study supports previous findings in hippocampal development showing
relatively larger bilateral hippocampal volumes in females compared to males and significant age-
related volumetric growth. The findings also provide novel evidence identifying hippocampal
regions that contribute to these overall differences. Specifically, age-related volumetric
enlargement of the hippocampus is accompanied by global linear surface expansion in the left and
right hippocampus, and nonlinear expansion of the head and tail primarily in the right hippocampal
surface. The larger adjusted hippocampal volumes in females may be localized to the lateral and
mesial surfaces bilaterally and could be further exaggerated by the faster maturation of the left
lateral surface observed in females compared to males. Results from this large, multi-site study
across infancy and early adulthood contributes to a growing compendium of evidence describing
dynamic developmental trajectories in structural hippocampal maturation that are characterized by
a complex interplay between age and sex. Future studies should aim to correlate changes in
memory with hippocampal shape in development to determine how functional modification is
reflected in structure.

91

3.6 References

Altemus KL, Lavenex P, Ishizuka N, Amaral DG. 2005. Morphological characteristics and
electrophysiological properties of CA1 pyramidal neurons in macaque monkeys.
Neuroscience. 136:741–756.
Aly M, Turk-Browne NB. 2017. Flexible weighting of diverse inputs makes hippocampal
function malleable. Neurosci Lett. [Epub ahea.
Amunts K, Kedo O, Kindler M, Pieperhoff P, Mohlberg H, Shah NJ, Habel U, Schneider F,
Zilles K. 2005. Cytoarchitectonic mapping of the human amygdala, hippocampal region and
entorhinal cortex: Intersubject variability and probability maps. Anat Embryol (Berl).
210:343–352.
Arnold SE, Trojanowski JQ. 1996. Human Fetal Hippocampal Development : and Neuronal
Morphologic Features. J Comp Neurol. 367:274–292.
Augustinack JC, Helmer K, Huber KE, Kakunoori S, Zöllei L, Fischl B, Kroenke C, Health O.
2010. Direct visualization of the perforant pathway in the human brain with ex vivo
diffusion tensor imaging. Front Hum Neurosci. 4:1–13.
Bannerman DM, Rawlins JNP, McHugh SB, Deacon RMJ, Yee BK, Bast T, Zhang WN,
Pothuizen HHJ, Feldon J. 2004. Regional dissociations within the hippocampus - Memory
and anxiety. Neurosci Biobehav Rev. 28:273–283.
Barnea-Goraly N, Frazier TW, Piacenza L, Minshew NJ, Keshavan MS, Reiss AL, Hardan AY.
2014. A preliminary longitudinal volumetric MRI study of amygdala and hippocampal
volumes in autism. Prog Neuro-Psychopharmacology Biol Psychiatry. 48:124–128.
Barth M, Breuer F, Koopmans PJ, Norris DG, Poser BA. 2016. Simultaneous multislice (SMS)
imaging techniques. Magn Reson Med. 75:63–81.
Bean LA, Ianov L, Foster TC. 2014. Estrogen receptors, the hippocampus, and memory.
Neuroscientist. 20:534–545.
Berman S, O’Neill J, Fears S, Bartzokis G, London ED. 2008. Abuse of amphetamines and
structural abnormalities in the brain. Ann N Y Acad Sci. 1141:195–220.
Bijanki KR, Hodis B, Brumm MC, Harlynn EL, McCormick LM. 2014. Hippocampal and left
subcallosal anterior cingulate atrophy in psychotic depression. PLoS One. 9:1–7.

92

Bowers JM, Waddell J, McCarthy MM. 2010. A developmental sex difference in hippocampal
neurogenesis is mediated by endogenous oestradiol. Biol Sex Differ. 1:8.
Bramen JE, Hranilovich JA, Dahl RE, Forbes EE, Chen J, Toga AW, Dinov ID, Worthman CM,
Sowell ER. 2011. Puberty influences medial temporal lobe and cortical gray matter
maturation differently in boys than girls matched for sexual maturity. Cereb Cortex.
21:636–646.
Bremner JD, Narayan M, Anderson ER, Staib LH, Miller HL, Charney DS. 2000. Hippocampal
volume reduction in major depression. Am J Psychiatry. 157:115–118.
Briellmann RS, Syngeniotis A, Jackson GD. 2001. Comparison of hippocampal volumetry at 1.5
tesla and at 3 tesla. Epilepsia. 42:1021–1024.
Burgess N, Maguire EA, O’Keefe J. 2002. The human hippocampus and spatial and episodic
memory. Neuron. 35:625–641.
Cao J, Worsley KJ. 1999. The detection of local shape changes via the geometry of Hotelling’s
T2 fields. Ann Stat. 27:925–942.
Catenoix H, Magnin M, Mauguière F, Ryvlin P. 2011. Evoked potential study of hippocampal
efferent projections in the human brain. Clin Neurophysiol. 122:2488–2497.
Chen J, Liu J, Calhoun VD, Arias-Vasquez A, Zwiers MP, Gupta CN, Franke B, Turner JA.
2014. Exploration of scanning effects in multi-site structural MRI studies. J Neurosci
Methods. 230:37–50.
Chua EF, Schacter DL, Rand-Giovannetti E, Sperling RA. 2007. Evidence for a Specific Role of
the Anterior Hippocampal Region in Successful Associative Encoding. Hippocampus.
17:1071–1080.
Cooke BM, Woolley CS. 2005. Gonadal hormone modulation of dendrites in the mammalian
CNS. J Neurobiol. 64:34–46.
Costafreda SG, Dinov ID, Tu Z, Shi Y, Liu CY, Kloszewska I, Mecocci P, Soininen H, Tsolaki
M, Vellas B, Wahlund LO, Spenger C, Toga AW, Lovestone S, Simmons A. 2011.
Automated hippocampal shape analysis predicts the onset of dementia in mild cognitive
impairment. Neuroimage. 56:212–219.
Dandolo LC, Schwabe L. 2018. Time-dependent memory transformation along the hippocampal
anterior–posterior axis. Nat Commun. 9:1205.

93

DeMaster D, Pathman T, Lee JK, Ghetti S. 2014. Structural development of the hippocampus
and episodic memory: Developmental differences along the anterior/posterior axis. Cereb
Cortex. 24:3036–3045.
Dennis C V., Suh LS, Rodriguez ML, Kril JJ, Sutherland GT. 2016. Human adult neurogenesis
across the ages: An immunohistochemical study. Neuropathol Appl Neurobiol. 1–18.
Dimsdale-Zucker HR, Ritchey M, Ekstrom AD, Yonelinas AP, Ranganath C. 2018. CA1 and
CA3 differentially support spontaneous retrieval of episodic contexts within human
hippocampal subfields. Nat Commun. 9.
Ding SL, Van Hoesen GW. 2015. Organization and detailed parcellation of human hippocampal
head and body regions based on a combined analysis of Cyto- and chemoarchitecture. J
Comp Neurol. 523:2233–2253.
Duvernoy H, Cattin F, Risold P-Y. 2013. The Human Hippocampus: Functional anatomy,
vascularization and serial sections with MRI. 4th Ed. ed. Berlin, Germany: Springer.
Eldridge LL, Engel SA, Zeineh MM, Bookheimer SY, Knowlton BJ. 2005. A Dissociation of
Encoding and Retrieval Processes in the Human Hippocampus. J Neurosci. 25:3280–3286.
Fanselow MS, Dong HW. 2010. Are the Dorsal and Ventral Hippocampus Functionally Distinct
Structures? Neuron. 65:7–19.
Fennema-Notestine C, Gamst AC, Quinn BT, Pacheco J, Jernigan TL, Thal L, Buckner R,
Killiany R, Blacker D, Dale AM, Fischl B, Dickerson B, Gollub RL. 2007. Feasibility of
multi-site clinical structural neuroimaging studies of aging using legacy data.
Neuroinformatics. 5:235–245.
Friston KJ. 1997. Testing for Anatomically Specified Regional Effects. Hum Brain Mapp.
5:133–136.
Ghetti S, Angelini L. 2008. The Development of Recollection and Familiarity in Childhood and
Adolescence : Evidence from the Dual-Process Signal Detection Model Author ( s ):
Simona Ghetti and Laura Angelini Published by : Wiley on behalf of the Society for
Research in Child Develop. Child Dev. 79:339–358.
Ghetti S, Lee J. 2011. Children’s episodic memory. Wiley Interdiscip Rev Cogn Sci. 2:365–373.
Giedd J, Vaituzis C, Hamburger S, Lange N, Rajapakse JC, Kaysen D, Vauss YC, Rapoport JL.
1996. Quantitative MRI of the Temoral Lobe, Amydala, and Hippocampus in Normal
Human Development: Ages 4-18 Years. J Comp Neurol. 366:223–230.

94

Giuliano A, Donatelli G, Cosottini M, Tosetti M, Retico A, Fantacci ME. 2017. Hippocampal
subfields at ultra high field MRI: An overview of segmentation and measurement methods.
Hippocampus. 27:481–494.
Gogtay N, Nugent TF, Herman DH, Ordonez A, Greenstein D, Hyashi KM, Clasen L, Toga AW,
Giedd JN, Rapoport JL, Thompson PM. 2006. Dynamic Mapping of Normal Human
Hippocampal Development. Hippocampus. 16:664–672.
Greve A, Evans CJ, Graham KS, Wilding EL. 2011. Functional specialisation in the
hippocampus and perirhinal cortex during the encoding of verbal associations.
Neuropsychologia. 49:2746–2754.
Guo Y, Wu G, Yap P-T, Jewells V, Lin W, Shen D. 2015. Segmentation of Infant Hippocampus
Using Common Feature Representations Learned for Multimodal Longitudinal Data. Med
Image Comput Comput Assist Interv. 9351:63–71.
Haukvik UK, Westlye LT, Mørch-Johnsen L, Jørgensen KN, Lange EH, Dale AM, Melle I,
Andreassen OA, Agartz I. 2015. In vivo hippocampal subfield volumes in schizophrenia
and bipolar disorder. Biol Psychiatry. 77:581–588.
Henke K, Weber B, Kneifel S, Wieser HG, Buck A. 1999. Human hippocampus associates
information in memory. Proc Natl Acad Sci U S A. 96:5884–5889.
Hill J, Inder T, Neil J, Dierker D, Harwell J, Van Essen D. 2010. Similar patterns of cortical
expansion during human development and evolution. Proc Natl Acad Sci U S A.
107:13135–13140.
Holland S, Schmithorst VJ, Wagner M, Lee G, Rajagopal A, Sroka M, Felicelli N, Rupert A,
Clark K, Toga A, Freund L. 2015. The C-MIND project: normative MRI and behavioral
data from children from birth to 18 years. In: Proceedings of the 21st Annumal Meeting of
the Organization for Human Brain Mapping (OHBM), Honolulu, HA.
Hueston CM, Cryan JF, Nolan YM. 2017. Stress and adolescent hippocampal neurogenesis: Diet
and exercise as cognitive modulators. Transl Psychiatry. 7:e1081-17.
Iglesias JE, Augustinack JC, Nguyen K, Player CM, Player A, Wright M, Roy N, Frosch MP,
McKee AC, Wald LL, Fischl B, Van Leemput K. 2015. A computational atlas of the
hippocampal formation using ex vivo, ultra-high resolution MRI: Application to adaptive
segmentation of in vivo MRI. Neuroimage. 115:117–137.

95

Insausti R, Cebada-Sánchez S, Marcos P. 2010. Postnatal development of the human
hippocampal formation. Adv Anat Embryol Cell Biol. 206:1–86.
Ji JX, Son JB, Rane SD. 2007. PULSAR: A MATLAB Toolbox for Parallel Magnetic Resonance
Imaging Using Array Coils and Multiple Channel Receivers. Concepts Magn Reson Part B
Magn Reson Eng. 31B:24–36.
Johnson SLM, Wang L, Alpert KI, Greenstein D, Clasen L, Lalonde F, Miller R, Paroport J,
Gogtay N. 2013. Hippocampal shape abnormalities of patients with childhoof-onset
schizophrenia and their unaffected siblings. J Am Acad Child Adolesc Psychiatry. 52:527–
536.
Jovicich J, Czanner S, Han X, Salat D, van der Kouwe A, Quinn B, Pacheco J, Albert M,
Killiany R, Blacker D, Maguire P, Rosas D, Makris N, Gollub R, Dale A, Dickerson BC,
Fischl B. 2009. MRI-derived measurements of human subcortical, ventricular and
intracranial brain volumes: Reliability effects of scan sessions, acquisition sequences, data
analyses, scanner upgrade, scanner vendors and field strengths. Neuroimage. 46:177–192.
Jovicich J, Marizzoni M, Sala-Llonch R, Bosch B, Bartrés-Faz D, Arnold J, Benninghoff J,
Wiltfang J, Roccatagliata L, Nobili F, Hensch T, Tränkner A, Schönknecht P, Leroy M,
Lopes R, Bordet R, Chanoine V, Ranjeva JP, Didic M, Gros-Dagnac H, Payoux P,
Zoccatelli G, Alessandrini F, Beltramello A, Bargalló N, Blin O, Frisoni GB. 2013. Brain
morphometry reproducibility in multi-center 3T MRI studies: A comparison of cross-
sectional and longitudinal segmentations. Neuroimage. 83:472–484.
Kahn I, Andrews-Hanna JR, Vincent JL, Snyder AZ, Buckner RL. 2008. Distinct Cortical
Anatomy Linked to Subregions of the Medial Temporal Lobe Revealed by Intrinsic
Functional Connectivity. J Neurophysiol. 100:129–139.
Kalmady SV, Shivakumar V, Arasappa R, Subramaniam A, Gautham S, Venkatasubramanian G,
Gangadhar BN. 2017. Clinical correlates of hippocampus volume and shape in
antipsychotic-naïve schizophrenia. Psychiatry Res - Neuroimaging. 263:93–102.
Keller TA, Just MA. 2016. NeuroImage Structural and functional neuroplasticity in human
learning of spatial routes. Neuroimage. 125:256–266.
Kerchner GA, Hess CP, Hammond-Rosenbluth KE, Xu D, Rabinovici GD, Kelley DAC,
Vigneron DB, Nelson SJ, Miller BL. 2010. Hippocampal CA1 apical neuropil atrophy in
mild Alzheimer disease visualized with 7-T MRI. Neurology. 75:1381–1387.

96

Knickmeyer R, Gouttard S. 2008. A structural MRI study of human brain development from
birth to 2 years. J Neurosci. 28:12176–12182.
Kornack DR, Rakic P. 1999. Continuation of neurogenesis in the hippocampus of the adult
macaque monkey. Proc Natl Acad Sci U S A. 96:5768–5773.
Lavenex P, Banta Lavenex P. 2013. Building hippocampal circuits to learn and remember:
Insights into the development of human memory. Behav Brain Res. 254:8–21.
Lavenex P, Banta Lavenex P, Amaral DG. 2007. Postnatal development of the primate
hippocampal formation. Dev Neurosci. 29:179–192.
Leal SL, Yassa MA. 2018. Integrating new findings and examining clinical applications of
pattern separation. Nat Neurosci. 21:163–173.
Libby LA, Ekstrom AD, Ragland JD, Ranganath C. 2012. Differential Connectivity of Perirhinal
and Parahippocampal Cortices within Human Hippocampal Subregions Revealed by High-
Resolution Functional Imaging. J Neurosci. 32:6550–6560.
Lin M, Fwu PT, Buss C, Davis EP, Head K, Muftuler LT, Sandman CA, Su M-Y. 2013.
Developmental changes in hippocampal shape among preadolescent children. Int J Dev
Neurosci. 31:473–481.
Lindgren L, Bergdahl J, Nyberg L. 2016. Longitudinal Evidence for Smaller Hippocampus
Volume as a Vulnerability Factor for Perceived Stress. Cereb Cortex. 26:3527–3533.
Machado-de-Sousa JP, De Lima Osório F, Jackowski AP, Bressan RA, Chagas MHN, Torro-
Alves N, DePaula ALD, Crippa JAS, Hallak JEC. 2014. Increased amygdalar and
hippocampal volumes in young adults with social anxiety. PLoS One. 9:1–5.
Mack ML, Love BC, Preston AR. 2017. Building concepts one episode at a time: The
hippocampus and concept formation. Neurosci Lett. Epub ahead.
Maguire EA, Burgess N, Donnett JG, Frackowiak RSJ, Frith CD, Okeefe J. 1998. Knowing
where and getting there. Science (80- ). 280:921–924.
Maguire EA, Gadian DG, Johnsrude IS, Good CD, Ashburner J, Frackowiak RSJ, Frith CD.
2000. Navigation-related structural change in the hippocampi of taxi drivers. Proc Natl
Acad Sci. 97:4398–4403.
Maguire EA, Woolett K, Spiers HJ. 2006. London Taxi Drivers and Bus Drivers: A Structural
MRI and Neuropsychological Analysis. Hippocampus. 16:1091–1101.

97

Maier S, Tebartz van Elst L, Beier D, Ebert D, Fangmeier T, Radtke M, Perlov E, Riedel A.
2015. Increased hippocampal volumes in adults with high functioning autism spectrum
disorder and an IQ>100: A manual morphometric study. Psychiatry Res - Neuroimaging.
234:152–155.
Maller JJ, Broadhouse K, Rush AJ, Gordon E, Koslow S, Grieve SM. 2017. Increased
hippocampal tail volume predicts depression status and remission to anti-depressant
medications in major depression. Nat Publ Gr. 1–8.
Manns JR, Hopkins RO, Squire LR. 2003. Semantic memory and the human hippocampus.
Neuron. 38:127–133.
McEwen BS, Milner TA. 2007. Hippocampal formation: Shedding light on the influence of sex
and stress on the brain. Brain Res Rev. 55:343–355.
Neufang S, Specht K, Hausmann M, Güntürkün O, Herpertz-Dahlmann B, Fink GR, Konrad K.
2009. Sex differences and the impact of steroid hormones on the developing human brain.
Cereb Cortex. 19:464–473.
Nordenskjöld R, Malmberg F, Larsson EM, Simmons A, Ahlström H, Johansson L, Kullberg J.
2015. Intracranial volume normalization methods: Considerations when investigating
gender differences in regional brain volume. Psychiat Res-Neuroim. 231:227–235.
Ofen N, Shing YL. 2013. From perception to memory: Changes in memory systems across the
lifespan. Neurosci Biobehav Rev. 37:2258–2267.
Østby Y, Tamnes CK, Fjell AM, Westlye LT, Due-Tønnessen P, Walhovd KB. 2009.
Heterogeneity in Subcortical Brain Development: A Structural Magnetic Resonance
Imaging Study of Brain Maturation from 8 to 30 Years. J Neurosci. 29:11772–11782.
Pantazatos SP, Talati A, Schneier FR, Hirsch J. 2014. Reduced Anterior Temporal and
Hippocampal Functional Connectivity During Face Processing Discriminates Individuals
with Social Anxiety Disorder from Healthy Controls and Panic Disorder, and Increases
Following Treatment. Neuropsychopharmacology. 39211:425–434.
Pardoe H, Pell GS, Abbott DF, Berg AT, Jackson GD. 2008. Multi-site voxel-based
morphometry: Methods and a feasibility demonstration with childhood absence epilepsy.
Neuroimage. 42:611–616.

98

Parekh MB, Rutt BK, Purcell R, Chen Y, Zeineh MM. 2015. Ultra-high resolution in-vivo 7 . 0 T
structural imaging of the human hippocampus reveals the endfolial pathway. Neuroimage.
112:1–6.
Patenaude B, Smith SM, Kennedy DN, Jenkinson M. 2011. A Bayesian model of shape and
appearance for subcortical brain segmentation. Neuroimage. 56:907–922.
Paul EW, Harrison J. 2004. The hippocampus in schizophrenia: a review of the
neuropathological evidence and its pathophysiological implications. Psychopharmacology
(Berl). 174:151–162.
Paus T, Keshavan M, Giedd JN. 2008. Why do many Psychiatric disorders emerge during
adolescence? Nat Rev Neurosci. 9:947–957.
Philip RCM, Dauvermann MR, Whalley HC, Baynham K, Lawrie SM, Stanfield AC. 2012. A
systematic review and meta-analysis of the fMRI investigation of autism spectrum
disorders. Neurosci Biobehav Rev. 36:901–942.
Poppenk J, Evensmoen HR, Moscovitch M, Nadel L. 2013. Long-axis specialization of the
human hippocampus. Trends Cogn Sci. 17:230–240.
Poppenk J, Moscovitch M. 2011. A hippocampal marker of recollection memory ability among
healthy young adults: Contributions of posterior and anterior segments. Neuron. 72:931–
937.
Powell TR, Murphy T, Lee SH, Duarte RRR, Lee HA, Smeeth D, Price J, Breen G, Thuret S.
2017. Inter-individual variation in genes governing human hippocampal progenitor
differentiation in vitro is associated with hippocampal volume in adulthood. Sci Rep.
7:15112.
Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. 1999. SENSE: Sensitivity encoding
for fast MRI. Magn Reson Med. 42:952–962.
Pujol N, Penadés R, Junqué C, Dinov I, Fu CHY, Catalán R, Ibarretxe-Bilbao N, Bargalló N,
Bernardo M, Toga A, Howard RJ, Costafreda SG. 2014. Hippocampal abnormalities and
age in chronic schizophrenia: Morphometric study across the adult lifespan. Br J Psychiatry.
205:369–375.
Rajah MN, Kromas M, Han JE, Pruessner JC. 2010. Group differences in anterior hippocampal
volume and in the retrieval of spatial and temporal context memory in healthy young versus
older adults. Neuropsychologia. 48:4020–4030.

99

Reagh ZM, Ranganath C. 2018. What does the functional organization of cortico-hippocampal
networks tell us about the functional organization of memory? Neurosci Lett.
Reagh ZM, Watabe J, Ly M, Murray E, Yassa MA. 2014. Dissociated Signals in Human Dentate
Gyrus and CA3 Predict Different Facets of Recognition Memory. J Neurosci. 34:13301–
13313.
Sabuncu MR, Ge T, Holmes AJ, Smoller JW, Buckner RL, Fischl B, Alzheimer’s Disease
Neuroimaging Initiative the ADN. 2016. Morphometricity as a measure of the
neuroanatomical signature of a trait. Proc Natl Acad Sci U S A. 113:E5749-56.
Saitoh O, Karns CM, Courchesne E. 2001. Development of the hippocampal formation from 2 to
42 years: MRI evidence of smaller area dentata in autism. Brain. 124:1317–1324.
Satpute AB, Mumford JA, Naliboff BD, Poldrack RA. 2012. Human anterior and posterior
hippocampus respond distinctly to state and trait anxiety. Emotion. 12:58–68.
Satterthwaite TD, Connolly JJ, Ruparel K, Calkins ME, Jackson C, Elliott MA, Roalf DR,
Hopsona R, Prabhakaran K, Behr M, Qiu H, Mentch FD, Chiavacci R, Sleiman PMA, Gur
RC, Hakonarson H, Gur RE. 2016. The Philadelphia Neurodevelopmental Cohort: A
publicly available resource for the study of normal and abnormal brain development in
youth. Neuroimage. 124:1115–1119.
Satterthwaite TD, Vandekar S, Wolf DH, Ruparel K, Roalf DR, Jackson C, Elliott MA, Bilker
WB, Calkins ME, Prabhakaran K, Davatzikos C, Hakonarson H, Gur RE, Gur RC. 2014.
Sex differences in the effect of puberty on hippocampal morphology. J Am Acad Child
Adolesc Psychiatry. 53:341–350.
Scharfman HE, MacLusky NJ. 2017. Sex differences in hippocampal area CA3 pyramidal cells.
J Neurosci Res. 95:563–575.
Segall JM, Turner JA, Van Erp TGM, White T, Bockholt HJ, Gollub RL, Ho BC, Magnotta V,
Jung RE, McCarley RW, Schulz SC, Lauriello J, Clark VP, Voyvodic JT, Diaz MT,
Calhoun VD. 2009. Voxel-based morphometric multisite collaborative study on
schizophrenia. Schizophr Bull. 35:82–95.
Sekeres MJ, Winocur G, Moscovitch M. 2018. The hippocampus and related neocortical
structures in memory transformation. Neurosci Lett. Accepted m.
Seress L. 1998. Neuronal connections, cell formation and cell migration in the perinatal human
hippocampal dentate gyrus. Ces Fysiol. 47:42–50.

100

Seress L, Ábrahám H. 2014. Pre- and postnatal morphological development of the human
hippocampal formation. In: Nelson CA,, Luciana M, editors. Handbook of Developmental
Cognitive Neuroscience. 2nd ed. MIT Press. p. 187–211.
Seress L, Mrzljak L. 1992. Postnatal development of mossy cells in the human dentate gyrus: A
light microscopic Golgi study. Hippocampus. 2:127–141.
Seress L, Ribak CE. 1995. Postnatal-Development of CA3 Pyramidal Neurons and Their
Afferents in the Ammons Horn of Rhesus-Monkeys. Hippocampus. 5:217–231.
Seri B, Garcia-Verdugo JM, McEwen BS, Alvarez-Buylla A. 2001. Astrocytes give rise to new
neurons in the adult mammalian hippocampus. J Neurosci. 21:7153–7160.
Setsompop K, Feinberg DA, Polimeni JR. 2016. Rapid brain MRI acquisition techniques at ultra-
high fields. NMR Biomed. 29:1198–1221.
Sheldon S, Levine B. 2016. The role of the hippocampus in memory and mental construction.
Ann N Y Acad Sci. 1369:76–92.
Shi F, Fan Y, Tang S, Gilmore JH, Lin W, Shen D. 2010. Neonatal brain image segmentation in
longitudinal MRI studies. Neuroimage. 49:391–400.
Shi F, Wang L, Wu G, Li G, Gilmore JH, Lin W, Shen D. 2014. Neonatal atlas construction
using sparse representation. Hum Brain Mapp. 35:4663–4677.
Shi F, Yap PT, Fan Y, Gilmore JH, Lin W, Shen D. 2010. Construction of multi-region-multi-
reference atlases for neonatal brain MRI segmentation. Neuroimage. 51:684–693.
Shi Y, Lai R, Krishna S, Sicotte N, Dinov I, Toga AW. 2008. Anisotropic Laplace-Beltrami
Eigenmaps: Bridging Reed graphs and skeletons. Proc IEEE Comput Soc Conf Comput Vis
Pattern Recognit. 2008:1–7.
Shi Y, Lai R, Morra JH, Dinov I, Thompson PM, Toga AW. 2010. Robust surface reconstruction
via Laplace-Beltrami eigen-projection and boundary deformation. IEEE Trans Med
Imaging. 29:2009–2022.
Shi Y, Lai R, Wang DJJ, Pelletier D, Mohr D, Sicotte N, Toga AW. 2014. Metric optimization
for surface analysis in the Laplace-Beltrami embedding space. IEEE Trans Med Imaging.
33:1447–1463.
Shi Y, Morra J, Thompson P, Toga A. 2009. Inverse-consistent surface mapping with Laplace-
Beltrami Eigen-Features. Comp Imag Vis. 21:467–478.

101

Shokouhi M, Barnes A, Suckling J, Moorhead TW, Brennan D, Job D, Lymer K, Dazzan P, Reis
Marques T, MacKay C, McKie S, Williams SC, Lawrie SM, Deakin B, Williams SR,
Condon B. 2011. Assessment of the impact of the scanner-related factors on brain
morphometry analysis with Brainvisa. BMC Med Imaging. 11:23.
Shuter B, Yeh IB, Graham S, Au C, Wang SC. 2008. Reproducibility of brain tissue volumes in
longitudinal studies: Effects of changes in signal-to-noise ratio and scanner software.
Neuroimage. 41:371–379.
Smith DM, Mizumori SJY. 2006. Hippocampal Place Cells, Context, and Episodic Memory.
Hippocampus. 16:816–729.
Smith SM, Jenkinson M, Woolrich MW, Beckmann CF, Behrens TEJ, Johansen-Berg H,
Bannister PR, De Luca M, Drobnjak I, Flitney DE, Niazy RK, Saunders J, Vickers J, Zhang
Y, De Stefano N, Brady JM, Matthews PM. 2004. Advances in functional and structural
MR image analysis and implementation as FSL. Neuroimage. 23:208–219.
Sorrells SF, Paredes MF, Cebrian-silla A, Sandoval K, Qi D, Kevin W, James D, Mayer S,
Chang J, Auguste KI, Chang EF, Gutierrez AJ, Oldham MC, Huang EJ, Garcia-verdugo
JM, Kriegstein AR, Gary W. 2018. Human hippocampal neurogenesis drops sharply in
children to undetectable levels in adults. Nature. 555:377–381.
Spalding KL, Bergmann O, Alkass K, Bernard S, Salehpour M, Huttner HB, Boström E,
Westerlund I, Vial C, Buchholz BA, Possnert G, Mash DC, Druid H, Frisén J. 2013.
Dynamics of Hippocampal Neurogenesis in Adult Humans. Cell. 153:1219–1227.
Stokes J, Kyle C, Arne D. Ekstrom. 2015. Complementary Roles of Human Hippocampal
Subfields in Differentiation and Integration of Spatial Context. J Cogn Neurosci. 27:546–
559.
Stonnington CM, Tan G, Klöppel S, Chu C, Draganski B, Jack CR, Chen K, Ashburner J,
Frackowiak RSJ. 2008. Interpreting scan data acquired from multiple scanners: A study
with Alzheimer’s disease. Neuroimage. 39:1180–1185.
Strange BA, Witter MP, Lein ES, Moser EI. 2014. Functional organization of the hippocampal
longitudinal axis. Nat Rev Neurosci. 15:655–669.

102

Sun J, Bonaguidi MA, Jun H, Guo JU, Sun GJ, Will B, Yang Z, Jang M-H, Song H, Ming G,
Christian KM. 2015. A septo-temporal molecular gradient of sfrp3 in the dentate gyrus
differentially regulates quiescent adult hippocampal neural stem cell activation. Mol Brain.
8:52.
Takao H, Hayashi N, Ohtomo K. 2014. Effects of study design in multi-scanner voxel-based
morphometry studies. Neuroimage. 84:133–140.
Tavares RM, Mendelsohn A, Grossman Y, Shapiro M, Trope Y, Schiller D, Tavares RM,
Mendelsohn A, Grossman Y, Williams CH, Shapiro M. 2015. A Map for Social Navigation
in the Human Brain Article A Map for Social Navigation in the Human Brain. Neuron.
87:231–243.
Thompson PM, Hayashi KM, De Zubicaray GI, Janke AL, Rose SE, Semple J, Hong MS,
Herman DH, Gravano D, Doddrell DM, Toga AW. 2004. Mapping hippocampal and
ventricular change in Alzheimer disease. Neuroimage. 22:1754–1766.
Toda T, Parylak SL, Linker SB, Gage FH, Gage FH. 2018. The role of adult hippocampal
neurogenesis in brain health and disease. Mol Psychiatry. Epub ahead.
Uematsu A, Matsui M, Tanaka C, Takahashi T, Noguchi K, Suzuki M, Nishijo H. 2012.
Developmental Trajectories of Amygdala and Hippocampus from Infancy to Early
Adulthood in Healthy Individuals. PLoS One. 7:e46970.
Voevodskaya O, Simmons A, Nordenskjöld R, Kullberg J, Ahlström H, Lind L, Wahlund L-O,
Larsson E-M, Westman E, Initiative ADN. 2014. The effects of intracranial volume
adjustment approaches on multiple regional MRI volumes in healthy aging and Alzheimer’s
disease. Front Aging Neurosci. 6:1–14.
Voineskos AN, Winterburn JL, Felsky D, Pipitone J, Rajji TK, Mulsant BH, Chakravarty MM.
2015. Hippocampal (subfield) volume and shape in relation to cognitive performance across
the adult lifespan. Hum Brain Mapp. 36:3020–3037.
Wiggins GC, Triantafyllou C, Potthast A, Reykowski A, Nittka M, Wald LL. 2006. 32-Channel
3 tesla receive-only phased-array head coil with soccer-ball element geometry. Magn Reson
Med. 56:216–223.
Wilke M, Schmithorst VJ, Holland SK. 2002. Assessment of spatial normalization of whole-
brain magnetic resonance images in children. Hum Brain Mapp. 17:48–60.

103

Wilke M, Schmithorst VJ, Holland SK. 2003. Normative pediatric brain data for spatial
normalization and segmentation differs from standard adult data. Magn Reson Med.
50:749–757.
Williams L-A, Gelman N, Picot PA, Lee DS, Ewing JR, Han VK, Thompson RT. 2005.
Neonatal Brain: Regional Variability of in Vivo MR Imaging Relaxation Rates at 3.0 T—
Initial Experience. Radiology. 235:595–603.
Winterburn JL, Pruessner JC, Chavez S, Schira MM, Lobaugh NJ, Voineskos AN, Chakravarty
MM. 2013. A novel in vivo atlas of human hippocampal subfields using high-resolution 3T
magnetic resonance imaging. Neuroimage. 74:254–265.
Wise PM. 2006. Estrogen therapy: Does it help or hurt the adult and aging brain? Insights
derived from animal models. Neuroscience. 138:831–835.
Wisse LEM, Gerritsen L, Zwanenburg JJM, Kuijf HJ, Luijten PR, Biessels GJ, Geerlings MI.
2012. Subfields of the hippocampal formation at 7T MRI: In vivo volumetric assessment.
Neuroimage. 61:1043–1049.
Woo C-W, Krishnan A, Wager TD. 2014. Cluster-extent based thresholding in fMRI analyses:
Pitfalls and recommendations. Neuroimage. 91:412–419.
Worsley KJ, Andermann M, Koulis T, MacDonald D, Evans AC. 1999. Detecting changes in
nonisotropic images. Hum Brain Mapp. 8:89–101.
Worsley KJ, Evans AC, Marrett S, Neelin P. 1992. A three-dimensional statistical analysis for
CBF activation studies in human brain. J Cereb Blood Flow Metab. 12:900–918.
Yoon U, Fonov VS, Perusse D, Evans AC. 2009. The effect of template choice on morphometric
analysis of pediatric brain data. Neuroimage. 45:769–777.
Yuen GS, McEwen BS, Akama KT. 2011. LIM kinase mediates estrogen action on the actin
depolymerization factor Cofilin. Brain Res. 1379:44–52.
Yurgelun-Todd DA, Killgore WDS, Cintron CB. 2003. Cognitive correlates of medial temporal
lobe development across adolescence: A magnetic resonance imaging study. Percept Mot
Skills. 96:3–17.

104

Yushkevich PA, Amaral RSC, Augustinack JC, Bender AR, Bernstein JD, Boccardi M,
Bocchetta M, Burggren AC, Carr VA, Chakravarty MM, Chetelat G, Daugherty AM,
Davachi L, Ding S-L, Ekstron A, Geerlings MI, Hassan A, Huang Y, Iglesias E, Joie R La,
Kerchner GA, LaRocque KF, Libby LA, Malykhin N, Mueller SG, Olsen RK, Palombo DJ,
Parekh MB, Pluta JB, Preston AR, Pruessner JC, Ranganath C, Raz N, Schlichtung M,
Schoemaker D, Singh S, Stark CEL, Suthana N, Tompary A, Turowski MM, Leemput K
Van, Wagner AD, Wang L, Winterburn JL, Wisse LEM, Yassa MA, Zeineh MM, (HSG)
for the HSG. 2015. Quantitative comparison of 21 protocols for labeling hippocampal
subfields and parahippocampal subregions in in vivo MRI: Towards a harmonized
segmentation protocol. Neuroimage. 111:526–541.
Yushkevich PA, Pluta JB, Wang H, Xie L, Ding SL, Gertje EC, Mancuso L, Kliot D, Das SR,
Wolk DA. 2015. Automated volumetry and regional thickness analysis of hippocampal
subfields and medial temporal cortical structures in mild cognitive impairment. Hum Brain
Mapp. 36:258–287.
Zeidman P, Lutti A, Maguire EA. 2015. Investigating the functions of subregions within anterior
hippocampus. Cortex. 73:240–256.
Zeineh MM, Palomero-Gallagher N, Axer M, Gräβel D, Goubran M, Wree A, Woods R, Amunts
K, Zilles K. 2016. Direct Visualization and Mapping of the Spatial Course of Fiber Tracts at
Microscopic Resolution in the Human Hippocampus. Cereb Cortex. 1–16.

105

3.7 Supplementary materials

Supplementary Table 1 MRI acquisition parameters

Study Site Scanner Model Scan Parameters
C-MIND Cincinnati Children's Hospital Philips Achieva
TR=8.1ms, TE=3.7, flip angle=8˚,
matrix size=256x160, voxel
size=1x1x1mm, acquisition time:
5:15min
PNC
Children's Hospital of
Philadelphia
Siemens
TrioTim
TR=1810ms, TE=3.5, flip angle=9˚,
matrix size=256x160, voxel
size=.9375x.9375x1mm, acquisition
time: 3:28min
PING
Massachusetts General
Hospital
Siemens
TrioTim
TR=2170ms, TE=4.33ms, flip
angle=7˚, matrix size=256x256, voxel
size: 1x1x1.2mm, acquisition time:
8:06min
Yale University
University of California,
Los Angeles
University of Hawaii
Weill Cornell Medical College
John Hopkins University
Philips Achieva
TR=6.8ms, TE=3.1, flip angle=8˚,
matrix size=256x240, voxel
size=1x1x1.2mm, acquisition time:
9:19.7min
Children's Hospital, Los
Angeles
University of California, San
Diego
GE Signa TR=8.1ms, TE=3.5ms, flip angle=8˚,
matrix size=256x192, voxel
size=.9375x.9375x1.2, acquisition
time: 8:05min
GE Discovery

106

Supplementary Figure 1: Main effect of age using shape analysis in subsetted populations.
(A) The results from Figure 4 were reproduced in the complete dataset, (B) after excluding infants
less than 3 years of age, and (C) after excluding children younger than 8.25 years of age in order
to determine if data collected from different studies confounded our results. Results do not change
significantly after removing infants, however quadratic age effects disappear after removing
children younger than 8.25 years of age. Map shows adjusted p-value for linear (blue) and
quadratic (red) expansion using BIC model selection.

107

Supplementary Figure 2: Main effect of sex in subsetted populations.
(A) The results from Figure 5a were reproduced in the complete dataset, (B) after excluding infants
less than 3 years of age, and (C) after excluding children younger than 8.25 years of age in order
to determine if data collected from different studies affected our results. The pattern of significant
sex differences does not noticeably change after removal of infants or children younger than 8.25
years of age. Map shows adjusted p-values for clusters showing significant Female > Male
thickness.

108

Supplementary Figure 3: Interaction between age and sex in subsetted populations in the left
hippocampus.
(A) The results from Figure 5b were reproduced in the complete dataset, (B) after excluding infants
less than 3 years of age, and (C) after excluding children younger than 8.25 years of age in order
to determine if data collected from different studies confounded our results. The location of the
significant cluster does not change after removing infants or children young than 8.25 years of age.
Map shows adjusted p-value for clusters showing significant Female > Male thickness rate of
change over time.

109

Chapter 4
The effect of BMI on hippocampal shape across childhood

4.1 Abstract

Childhood obesity is associated with negative physiological and cognitive health outcomes. The
hippocampus is a diverse subcortical structure involved in learned feeding behaviors and energy
regulation, and research has shown that the hippocampus is vulnerable to the effects of excess
adiposity. Previous studies have demonstrated reduced hippocampal volume in overweight and
obese children, however it is unclear if certain sub-regions are selectively affected. The purpose
of this study was to determine the relationship between BMI z-score and regional hippocampal
surface morphology in a large cross-sectional multi-study cohort of 782 children and adolescents
(2.3 – 18.8 years) using a novel intrinsic shape analysis approach. We show significant inward
surface deformations with increasing BMI z-score in the left anterior hippocampus in 4 clusters
corresponding to putative CA1 and subiculum subfields. These results demonstrate the progressive
influence of excess weight on hippocampal structure that precede the development of obesity-
related atrophy and may represent potential therapeutic targets to promote potential recovery.

4.2 Introduction

The prevalence of childhood obesity has risen dramatically over the past 3 decades, with
recent estimates that more than 30% of children are overweight or obese (Ogden et al. 2014). In
particular, childhood obesity is increasing in communities with low socioeconomic status
(Frederick et al. 2014). While a gradual increase in adiposity is expected from childhood through
adolescence (Cole and Lobstein 2012), persistent childhood obesity leads to adverse
neuropsychological and health issues (Carnell et al. 2012). Obesity is an important factor for the
development of cardiovascular disease, Type 2 diabetes, stroke, and cancer (Haslam and James
2005; Pinhas-Hamel and Zeitler 2005; Haines et al. 2007). In addition to the increased risk of
physiological complications later in life, childhood obesity is also associated with a higher
prevalence of depression, psychiatric disorders, and cognitive issues (Braet et al. 1997; Erermis et
al. 2004; Allen et al. 2006). Specifically, a growing body of research shows that obesity can

110

increase the risk of developing memory problems in children (Cheke et al. 2016) and, over time,
central nervous system pathologies such as dementia and Alzheimer’s disease (Fitzpatrick et al.
2009; Mayeux and Stern 2012). Due to the high prevalence of and the myriad health issues
associated with childhood obesity, further research is needed to establish how these changes affect
the developing brain.
The hippocampal formation is a subcortical structure within the medial temporal lobe that
is critical for memory and learning processes (Sheldon and Levine 2016). Additionally, the
hippocampus is implicated in a number of metabolic functions, including appetitive, ingestive, and
learned eating behaviors (Davidson et al. 2007; Benoit et al. 2010; Kanoski 2012; Kanoski and
Grill 2017). The hippocampus is innervated by the hypothalamus, amygdala, and thalamus, which
form networks critical for energy metabolism and weight management (Davidson and Jarrard
1993; Davidson et al. 2005), and has receptor expression for leptin, an important metabolic
hormone (Van Doorn et al. 2017). Previous studies in children show obesity is associated with
deficits in hippocampal-dependent processes (Kanoski 2012; Miller and Spencer 2014; Khan et al.
2015; Cheke et al. 2016), and evidence in rodents suggest this impairment is unique to juveniles
(Boitard et al. 2012). Because hippocampal maturation is a protracted process (Giedd et al. 1996;
Insausti et al. 2010), it is important to study the unique vulnerabilities in this structure due to
childhood adiposity during critical periods of development.
Previous evidence suggests that these cognitive deficits observed in childhood obesity may
be accompanied by structural changes in the hippocampus. In rodent models, obesity in juveniles
leads to decreased neurogenesis (Boitard et al. 2012), synaptic stripping (Hao et al. 2016) and
neuroinflammatory processes (Miller and Spencer 2014; Guillemot-Legris and Muccioli 2017).
Hippocampal pathology subsequently manifests as selective changes to cornu ammonis 1 (CA1)
and CA3 subfields in the hippocampus (Sack et al. 2017). Human neuroimaging studies in vivo
have demonstrated that childhood obesity is associated with reduced hippocampal volume
(Chaddock et al. 2010; Bauer et al. 2015; Mestre et al. 2017; Nouwen et al. 2017), however, few
studies have explored the progressive influence of weight gain on hippocampal structure across
development.
The hippocampus is structurally and functionally heterogeneous (Poppenk et al. 2013;
Strange et al. 2014) and it is likely that whole volumetric approaches obscure localized changes in
the hippocampus associated with excess adiposity. Therefore, methods with enhanced regional

111

specificity to adiposity-related changes can provide insight into vulnerable subregions and
processes. Shape analysis provides a mechanism to detect regional deformations in surface
topology from triangulated meshes generated from hippocampal volumes (Shi et al. 2009, 2014).
Radial distance surface features provide vertex-wise measures of the hippocampal cross-sectional
radius and represents local thickness estimates (Thompson et al. 2004; Shi et al. 2009). Previous
studies have shown that global hippocampal shape features may be a more informative biomarker
for memory performance compared to volumetry (Voineskos et al. 2015).
The purpose of this study is to characterize how adiposity and weight influence regional
hippocampal structure in a large multi-study cohort of 782 children and adolescents between 2 and
20 years of age. Body mass index (BMI), a quantity derived from body weight and height, is a
recommended index for classifying adiposity in children and adolescents (Cole et al. 2000a) and
is directly connected to the development of obesity (Ogden et al. 2016). Using shape analysis, we
sought to characterize the relationship between BMI and hippocampal shape to identify if
subregions within the hippocampus are selectively vulnerable to the effects of adiposity during
critical developmental periods. These findings could contribute to the current knowledge regarding
the influence of global metabolic processes on developing brain structures, which may have
significant implications for lifelong cognitive function

4.3 Materials and Methods

Subjects

Cross-sectional T1-weighted MRI images and demographic data were obtained from 4
neuroimaging databases of typical child development and combined into a single dataset to
increase our power to detect hippocampal differences due to adiposity. These databases include
the publicly available Philadelphia Neurodevelopmental Cohort (PNC) research initiative and the
Cincinnati MR Imaging of NeuroDevelopment (C-MIND) data repository. We additionally used
data from a study on typically developing children who were exposed to gestational diabetes in
utero and controls (GDM) and data from a study on structural brain development in dyslexia
(DYS).

112

C-MIND dataset

Data from 77 typically developing participants were considered from the C-MIND study.
Structural MRI images were acquired with the same protocol on a 3T Philips Achieva at Cincinnati
Children’s Hospital (Holland et al. 2015). All MRI images were manually checked for data quality
and scans with excessive motion were discarded. Of the subjects considered, 3 were excluded due
to poor data quality, resulting in 74 C-MIND subjects included in the present analysis (37 female,
range: 2.3-18.8 years, M=9.2, SD=4.5 years). Demographics of the subjects included is presented
in Table 4.1.

PNC dataset

Data from 590 typically developing children and adolescents were acquired by the PNC
study at Children’s Hospital of Philadelphia (Satterthwaite et al. 2016). Scans were acquired with
the same protocol on a 3T Siemens Tim Trio whole-body MRI. Of the subjects considered for the
study, 46 subjects were excluded due to the quality control issues stated above, resulting in 590
PNC subjects in the present analysis (306 female, range: 8.33-20 years, M=14.0, SD=3.1 years)
(Table 1).

DYS dataset

We acquired structural T1-weighted MRI data for a study on developmental dyslexia and
included 33 cognitively normal controls from that study in the present analyses. Data was collected
through the Connectivity and Network Development group in the Institute for Neuroimaging and
Informatics at the University of Southern California (USC) and participants were required to be
right-handed, English monolingual, and have a FSIQ≥80. Structural MRI data were acquired with
the same protocol on a Siemens MAGNETOM Prisma
fit
3 Tesla MRI scanner (Siemens Medical
Systems) with a 24-channel phased array coil at USC and all data was checked for quality (16
female, range: 2.3-18.8 years, M=10.2, SD=1.7 years) (Table 4.1).

113

GDM dataset

Structural neuroimaging data from 97 typically developing children were considered from
a study exploring the effects of in utero exposure to gestational diabetes mellitus (GDM) in the
developing brain during childhood and adolescence. Children in this cohort were born at Kaiser
Permanente Southern California (KPSC), a large health care organization that uses an integrated
electronic medical record (EMR) system. Children were excluded if they had a history of
premature birth (<37 weeks of gestation), congenital anomalies, neurological, psychiatric, or other
significant medical disorders, used medications known to alter metabolism (e.g., glucocorticoids),
contraindications to MRI or were left-handed. Each participating Institutional Review Board
approved this study (University of Southern California (USC) #HS-14-00034, KPSC #10282).
Participants’ parents gave written informed consent. Children provided written informed assent.

For the GDM cohort, T1-weighted images were acquired with the same protocol on a
Siemens MAGNETOM Prisma
fit
3 Tesla MRI scanner (Siemens Medical Systems) with a 24-
channel phased array coil. Of the subjects considered, 8 were discarded due to poor neuroimaging
data quality, resulting in 85 subjects (47 female, range: 7.3-11.2 years, M=8.3, SD=.8 years) (Table
4.1). There were 48 cognitively normal children who were exposed to GDM and 37 unexposed
controls included in the study.

Table 4.1 Study demographics for multi-site dataset
Age BMI z-score
Study N Mean SD Range Mean SD Range
CMN 74 (37 F) 9.19 4.45 2.3 - 18.8 0.36 0.97 -1.7 - 2.32
DYS 33 (16 F) 10.21 1.66 7.5 - 12.92 0.56 1.41 -2.04 - 2.54
GDM 85 (47 F) 8.34 0.84 7.33 - 11.23 0.7 1.11 -1.82 - 2.54
PNC 590 (306 F) 14.02 3.05 8.33 - 20 0.61 1.12 -2.53 - 2.76
Total 782 (406 F) 12.78 3.72 2.33 - 20 0.6 1.12 -2.53 - 2.76
SD, standard deviation; F, female

In total, 782 children and adolescents between the ages of 2.3 and 20 years of age (406
female, M=12.8, SD=3.7 years) were included from the C-MIND, PNC, DYS, and GDM datasets.

114

Participants had no self-reported history of neurological issues or head trauma. Weight and height
were also recorded during neuroimaging sessions in the above studies. Body mass index (BMI)
z-score, which adjusts for age and sex, was calculated using a BMI calculator out of the Children’s
Hospital of Philadelphia Research Institute based on the Centers for Disease Control and
Prevention’s Pediatric BMI z-score calculator, which is appropriate to use in children between the
ages of 2 and 20 years (https://zscore.research.chop.edu/).

MRI Acquisition and Image Processing

In all datasets, structural neuroimaging was acquired with three-dimensional magnetization
prepared, rapid-acquisition gradient-echo (MPRAGE) T1-weighted sequences on 3T MRI
scanners with slightly different parameters used in each study (Supplemental Table 4.1). Because
voxel sizes differed between datasets, all T1w data was first resampled to 1 mm isotropic.
FSL’s anatomical processing pipeline was used for image processing and includes bias-
field correction, registration to standard MNI space, brain extraction and tissue-based
segmentation (Smith et al. 2004). Bilateral hippocampal volumes were segmented from the T1-
weighted image using FSL’s model-based segmentation tool, FIRST (Patenaude et al. 2011), and
were visually inspected for anatomical accuracy. Intra-cranial volume (ICV) was computed per
subject in cm
3
using the number of voxels contained within the skill.

Hippocampal Mapping/Shape Analysis

Shape analysis was performed using the fully-automated Metric Optimization for
Computational Anatomy (MOCA) software developed by Shi and colleagues (Shi et al. 2014) and
described previously (Lynch et al. 2018). MOCA performs intrinsic surface mapping using
Laplace-Beltrami (LB) eigen-functions, which computes the conformal maps directly between
anatomical surfaces. LB eigen-functions are invariant to isometry and result in robust preservation
of topology (Shi et al. 2010). Each segmented hippocampal volume was first converted to a
triangulated mesh in native space by iteratively updating vertices using surface deformation and
outlier detection, resulting in robust preservation of topology while avoiding shrinkage (Shi et al.
2010). In order to probe for regional changes in hippocampal structure related to adiposity, local

115

radial distance (RD) measures were computed using the Reeb graph of the first LB eigen-function
(Shi et al. 2008, 2009). This feature reflects local hippocampal thickness and is defined as the
shortest distance from each vertex to the longitudinal core of the hippocampus.
Individual meshes were first registered to a common space. From all the surfaces, we
generated the average surface as the geometric representation of the group-wise atlas by
minimizing the spectral l2-distance to all other surfaces. The atlases were then remeshed with 2000
vertices per atlas using SurfStat implemented in Matlab (www.math.mcgill.ca/keith/surfstat). The
features of each subject surface were directly mapped onto the average mesh by pulling back the
average mesh structure onto each subject mesh using linear interpolation, resulting in one-to-one
correspondence for statistical analyses.

Statistical Analysis

In order to determine the effect of adiposity on hippocampal structure using shape analysis,
general linear models (GLMs) were applied to each vertex in the group-wise atlas to test if radial
distance is significantly associated with BMI z-score, while controlling for age, sex, ICV and
study. Previous studies have demonstrated that the effect of scanner variability can be mitigated
with inclusion of study as covariates (Fennema-Notestine et al. 2007; Pardoe et al. 2008; Chen et
al. 2014; Takao et al. 2014). Because vertices in surface maps are spatially correlated, multiple
comparison techniques, such as Bonferroni correction, are not appropriate for shape analysis.
Random field theory (RFT) improves upon this by considering both the peaks and spatial extent
of smoothed statistical parameter maps using Gaussian random fields (Cao and Worsley 1999;
Worsley et al. 1999). The Euler characteristic estimates the corrected p-value at a given cluster
level (Friston 1997; Woo et al. 2014). This property is derived from the vertex-wise significance
and the number of resolution elements (resels) in the image, which describes the search volume as
a function of the image FWHM smoothness (Worsley et al. 1992). A supra-threshold cluster level
of p<.001 and a set level threshold of p<.05 was used for the height and spatial extent thresholds,
respectively. Region of interest (ROI) post hoc analyses were then performed on significant
clusters to better identify the relationship between adiposity and regional hippocampal thickness.

116

Figure 4.1 Regional hippocampal shape changes with BMI z-score.
(A) T-value maps of the linear relationship between BMI z-score and vertex-wise radial thickness
while controlling for age, age
2
, sex, ICV, and study on bilateral hippocampal surfaces. Warm
colors show hippocampal surface expansion with increasing BMI z-score and cool colors show
hippocampal surface contraction with increasing BMI z-score. (B) FDR-corrected p-value maps
showing clusters with radial distance that is significantly and negatively associated with BMI z-
score of the left hippocampus. On the superior surface, clusters located in the medial head ( β=-
.052, F(7,773)=19.42, corrected p<.008, adjusted R
2
=.159), and lateral head ( β=-.004,
F(8,773)=.088, corrected p<.026, adjusted R
2
=.079) reached significance. On the inferior surface,
clusters located in the lateral head ( β=-.048, F(8,773)=10.14, corrected p<.021, adjusted R
2
=.086),
and lateral body, ( β=-.05, F(8,773)=25.96, corrected p<.037, adjusted R
2
=.204) reached
significance.
4.4 Results

The children and adolescents included in this multi-study dataset had a mean age of 12.8
years (SD=3.7) and a mean BMI z-score of 0.6 (SD=1.1). According to the International Obesity

117

Task Force classification, the criteria to be classified as overweight/obese is with a BMI z-score
greater than 1.19 in girls and greater than 1.3 in boys. Using this criteria, 133 girls and 120 boys
qualify as overweight/obese in the present study (Cole et al. 2000b)
BMI z-score was not significantly associated with hippocampal volume bilaterally after
controlling for age, sex, ICV, and study, left: b=-.02, t(772)=-1.61, p=.11, right: b=.002,
t(772)=1.64, p=.10, however hippocampal volume was unsurprisingly significantly associated
with age, ICV and study.
Shape analysis results yield a significant relationship between radial distance and BMI z-
score in 4 clusters on the inferior and superior head of the left hippocampal surface (Figure 4.1).
In all clusters, radial distance was negatively associated with BMI z-score. BMI z-score was not
significantly associated with radial distance in the right hippocampus, however a small region in
the medial head trended toward significance.

4.5 Discussion

The purpose of this study was to examine the relationship between BMI and hippocampal
shape in a large cross-sectional cohort of healthy children and adolescents. The hippocampus has
been shown to be important in feeding behaviors (Davidson et al. 2007; Kanoski 2012; Kanoski
and Grill 2017) and previous studies have demonstrated hippocampal pathology associated with
excess weight gain (Bauer et al. 2015; Mestre et al. 2017; Nouwen et al. 2017). In the present
study, we demonstrate BMI z-score was negatively associated with radial distance across a range
of adiposity levels within 4 clusters of the left anterior hippocampus after controlling for sex, ICV,
study, and the quadratic effects of age.
Previous studies have shown that overweight and obese children have reduced left
hippocampal volumes compared to normal weight children (Bauer et al. 2015; Mestre et al. 2017).
In the present study, BMI z-score was not significantly associated with whole hippocampal
volume, which suggests that the observed inward surface deformations in the left hippocampus
made a subtle contribution to the overall structure. These results demonstrate that precursors of
hippocampal damage may be observed before the development of obesity. This is supported by
previous studies demonstrating hippocampal differences among normal weight children and
adults. In a longitudinal study by (Hashimoto et al. 2015), maturational expansion of the medial

118

temporal lobe gray matter is slowed in those with large BMI changes, suggesting that the brain is
sensitive to metabolic processes during developmental periods. Moreover, in adults, consumption
of a western diet is associated with smaller left hippocampal volumes (Jacka et al. 2015). It is
hypothesized that the hippocampus is particularly vulnerable to the effects of excess weight gain
due to the high energy demand, elevated cellular plasticity, and selective impairment of the blood-
brain barrier of the structure (Kanoski et al. 2010; Davidson et al. 2013; Williamson and Bilbo
2013; Hargrave et al. 2016).
A functional gradient exists along the anterior-posterior long axis of the human
hippocampus (Poppenk et al. 2013), and the anterior hippocampus is preferentially involved in
associative learning (Chua et al. 2007; Sheldon and Levine 2016) and processing global memory
features during consolidation (Sekeres et al. 2018). Feeding behaviors are heavily influenced by
learned associations between food-related sensory cues and interoceptive postingestive signals
(Kanoski and Grill 2017). Therefore, the anterior hippocampus may play a role in the conditioned
learning of rewarding or negative eating preferences.
The putative locations of the anterior clusters that demonstrate a negative relationship
between hippocampal thickness and BMI z-score are the CA1 and subiculum subfields (Insausti
et al. 2010). The importance of these regions for feeding behaviors can be demonstrated from
rodent studies, where the ventral hippocampus is analogous to the anterior hippocampus in
humans. The ventral CA1 and subiculum are involved in external contextual learning (Wilkerson
and Levin 1999; Kanoski and Davidson 2011; Beer et al. 2014; Keinath et al. 2014) and receive
indirect input from the olfactory bulb (Fanselow and Dong 2010) to process olfactory contextual
cues for learned associations between flavor and post-ingestive consequences (Keinath et al.
2014). The ventral CA1 and subiculum also processes internal signals related to reward incentives
(Gasbarri, Packard, et al. 1994; Gasbarri, Verney, et al. 1994), visceral energy status (Rozin et al.
1998; Higgs 2002), and emotion (Petrovich et al. 2001). Interoceptive information is transmitted
from the ventral hippocampus to regions important for feeding behaviors, such as the lateral
hypothalamic area (Stanley et al. 1996; Berthoud and Münzberg 2011; Hahn and Swanson 2012;
Brown et al. 2015). The ventral CA1 and subiculum also have projections to the medial prefrontal
cortex (mPFC) involved in hedonic feeding behaviors (Ishikawa and Nakamura 2006; Cenquizca
and Swanson 2007). It is proposed that the ventral hippocampus is involved in the motivational or
emotional components of overeating (Fanselow and Dong 2010), as evidenced from a study where

119

food intake was suppressed preferentially following administration of the hunger suppressive
hormone leptin in the ventral hippocampus (Kanoski and Davidson 2011).
The relationship between adiposity and hippocampal subfield size has not been explored
in human populations, but evidence from the rodent literature suggests a calorically dense diet is
associated with selective vulnerability in the hippocampus. In a combined histological and MRI
study in rodents, the acute effects of a western diet led to a significant increase in CA1 and CA3
subfield volumes unaccompanied by an increase in neurogenesis (Sack et al. 2017). While the
acute effects of poor diet may result in a transient increase in subfield volumes, it is possible that
the cumulative effects of chronic behaviors contributing to high BMI, such as overeating and
physical inactivity, may lead to hippocampal atrophy over time.
While the precise mechanisms by which increased weight gain influences regional
hippocampal structure over time are unknown, there is converging evidence that diet-induced
impairments are a result of neuroinflammatory processes (Shefer et al. 2013; Miller and Spencer
2014; Guillemot-Legris and Muccioli 2017). These processes are associated with increased
permeability of the blood-brain barrier (Banks et al. 2006; Kanoski et al. 2010), reduced
neurogenesis (Boitard et al. 2012), and synaptic stripping (Hao et al. 2016). Weight gain and
obesity can also result in chronic hypothamaic-pituitary-adrenal (HPA) axis activation (Spencer
and Tilbrook 2011), where sustained exposure to glucocorticoids results in hippocampal tissue loss
and reduced synaptic plasticity (Sapolsky 1985; Woolley et al. 1990; Kerr et al. 1991; Magariños
and McEwen 1995). It is not clear if the observed shape changes with increasing BMI z-score is
reflective of pathology, but it is possible that these regions that may be selectively vulnerable to
adiposity-induced inflammation observed in the development of clinical obesity. Future studies
should explore the nature of these gradual structural changes.
Childhood and adolescence are periods of rapid structural and functional maturation and
damage to the developing mind can lead to long-term consequences. The hippocampus undergoes
rapid and dynamic structural development throughout childhood and adolescence (Herting et al.
2018; Lynch et al. 2018) and neurogenesis continues through adulthood (Gómez and Edgin 2016).
Damage to the developing hippocampus could therefore derail normative developmental
trajectories, resulting in long lasting cognitive deficits and predisposition to metabolic and
cardiovascular disease. Given the present results, it is crucial to identify neurological processes
that may be vulnerable to the early effects of excess weight gain and promote the potential recovery

120

of hippocampal structure and function through early interventions with healthy habits such as
physical exercise (Herting and Nagel 2012).
A strength of our study is the large sample size used to identify the relationship between
adiposity and hippocampal structure across a large range of BMI z-scores. Our study adds to the
growing body of literature demonstrating the effect of metabolic processes on the hippocampus in
childhood (Bauer et al. 2015; Mestre et al. 2017; Nouwen et al. 2017). Additionally, shape analysis
provides the ability to detect localized changes in hippocampal morphology and provides enhanced
specificity over traditional volumetric approaches. However, the cross-sectional design of the
study limits the causal implications of BMI z-score on hippocampal structure. While previous
studies on individuals with obesity present with hippocampal-dependent memory impairments, it
is unclear if the observed shape changes with increasing BMI z-score are accompanied by
cognitive deficits due to the lack of neurocognitive measures. Lastly, while BMI is an acceptable
and readily available anthropometric measure for the clinical diagnosis of obesity (Cole et al.
2000a; Jean-Philippe Bastard et al. 2006; Krebs et al. 2007; Cole and Lobstein 2012), it is not the
most accurate quantification of body fat and lean tissue mass and has inconsistent associations
with body composition variables. For example, BMI z-score is a strong predictor for total fat mass
in children, however it is a weak predictor of percent body fat (Vanderwall et al. 2017). Future
studies should include additional measures that provide a more accurate profile of excess body fat
relative to body weight such as waist-to-hip ratio, waist-to-height ratio, insulin sensitivity, and
percent body fat (Barreira et al. 2013; Nambiar et al. 2013; Bacopoulou et al. 2015; Khan et al.
2015; Wicklow et al. 2015; Chula de Castro et al. 2018). However the ease of BMI acquisition
related to obesity in a large multi-study context enables the power to detect regional structural
changes across a range of body weights.
In conclusion, this study shows that increasing BMI z-score is negatively associated with
cross-sectional thickness in the left anterior hippocampus in a large cohort of children and
adolescents. Previous studies have explored hippocampal volumetric differences in children with
obesity and normal weight, however the present study is the first to demonstrate localized shape
changes across a range of adiposity levels. Hippocampal atrophy was localized to putative CA1
and subiculum subfields, suggesting hippocampal subregions may present a selective vulnerability
to metabolic changes that precede the development of obesity and its associated pathologies.

121

Longitudinal follow-up studies may provide further insight into the causal role of these processes
and identify therapies that could potentially reverse these structural changes.

4.6 References

Allen KL, Byrne SM, Blair EM, Davis EA. 2006. Why do some overweight children experience
psychological problems? the role of weight and shape concern. Int J Pediatr Obes. 1:239–
247.
Bacopoulou F, Efthymiou V, Landis G, Rentoumis A, Chrousos GP. 2015. Waist circumference,
waist-to-hip ratio and waist-to-height ratio reference percentiles for abdominal obesity
among Greek adolescents. BMC Pediatr. 15:1–9.
Banks WA, Farr SA, Morley JE. 2006. The effects of high fat diets on the blood-brain barrier
transport of leptin: Failure or adaptation? Physiol Behav. 88:244–248.
Barreira T, Staiano A, Katzmarzyk P. 2013. Validity Assessment of a Portable Bioimpedance
Scale to Estimate Body Fat Percentage in White and African American Children and
Adolescents. Pediatr Obes. 8:e29–e32.
Bauer CCC, Moreno B, González-Santos L, Concha L, Barquera S, Barrios FA. 2015. Child
overweight and obesity are associated with reduced executive cognitive performance and
brain alterations: A magnetic resonance imaging study in Mexican children. Pediatr Obes.
10:196–204.
Beer Z, Chwiesko C, Sauvage MM. 2014. Processing of spatial and non-spatial information
reveals functional homogeneity along the dorso-ventral axis of CA3, but not CA1.
Neurobiol Learn Mem. 111:56–64.
Benoit SC, Davis JF, Davidson TL. 2010. Learned and cognitive controls of food intake. Brain
Res. 1350:71–76.
Berthoud HR, Münzberg H. 2011. The lateral hypothalamus as integrator of metabolic and
environmental needs: From electrical self-stimulation to opto-genetics. Physiol Behav.
104:29–39.
Boitard C, Etchamendy N, Sauvant J, Aubert A, Tronel S, Marighetto A, Layé S, Ferreira G.
2012. Juvenile, but not adult exposure to high-fat diet impairs relational memory and
hippocampal neurogenesis in mice. Hippocampus. 22:2095–2100.

122

Braet C, Mervielde I, Vandereycken W. 1997. Psychological aspects of childhood obesity: A
controlled study in a clinical and nonclinical sample. J Pediatr Psychol. 22:59–71.
Brown JA, Woodworth HL, Leinninger GM. 2015. To ingest or rest? Specialized roles of lateral
hypothalamic area neurons in coordinating energy balance. Front Syst Neurosci. 9:1–25.
Cao J, Worsley KJ. 1999. The detection of local shape changes via the geometry of Hotelling’s
T2 fields. Ann Stat. 27:925–942.
Carnell S, Gibson C, Benson L, Ochner CN, Geliebter A. 2012. Neuroimaging and obesity:
Current knowledge and future directions. Obes Rev. 13:43–56.
Cenquizca LA, Swanson LW. 2007. Spatial organization of direct hippocampal field CA1 axonal
projections to the rest of the cerebral cortex. Brain Res Rev. 56:1–26.
Chaddock L, Erickson KI, Prakash RS, Kim JS, Voss MW, Vanpatter M, Pontifex MB, Raine
LB, Konkel A, Hillman CH, Cohen NJ, Kramer AF. 2010. A neuroimaging investigation of
the association between aerobic fitness, hippocampal volume, and memory performance in
preadolescent children. Brain Res. 1358:172–183.
Cheke LG, Simons JS, Clayton NS. 2016. Higher body mass index is associated with episodic
memory deficits in young adults. Q J Exp Psychol. 69:2305–2316.
Chen J, Liu J, Calhoun VD, Arias-Vasquez A, Zwiers MP, Gupta CN, Franke B, Turner JA.
2014. Exploration of scanning effects in multi-site structural MRI studies. J Neurosci
Methods. 230:37–50.
Chua EF, Schacter DL, Rand-Giovannetti E, Sperling RA. 2007. Evidence for a Specific Role of
the Anterior Hippocampal Region in Successful Associative Encoding. Hippocampus.
17:1071–1080.
Chula de Castro JA, Lima TR de, Silva DAS. 2018. Body composition estimation in children and
adolescents by bioelectrical impedance analysis: A systematic review. J Bodyw Mov Ther.
22:134–146.
Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. 2000a. Establishing a standard definition for child
overweight and obesity worldwide: international survey. BMJ. 320:1–6.
Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. 2000b. Establishing a standard definition for child
overweight and obesity worldwide: international survey. BMJ. 320:1240–1243.
Cole TJ, Lobstein T. 2012. Extended international (IOTF) body mass index cut-offs for thinness,
overweight and obesity. Pediatr Obes. 7:284–294.

123

Davidson TL, Hargrave SL, Swithers SE, Sample CH, Fu X, Kinzig KP, Zheng W. 2013. Inter-
relationships among diet, obesity and hippocampal-dependent cognitive function.
Neuroscience. 253:110–122.
Davidson TL, Jarrard LE. 1993. A role for hippocampus in the utilization of hunger signals.
Behav Neural Biol. 59:167–171.
Davidson TL, Kanoski SE, Schier LA, Clegg DJ, Benoit SC. 2007. A potential role for the
hippocampus in energy intake and body weight regulation. Curr Opin Pharmacol. 7:613–
616.
Davidson TL, Kanoski SE, Walls EK, Jarrard LE. 2005. Memory inhibition and energy
regulation. Physiol Behav. 86:731–746.
Erermis S, Cetin N, Tamar M, Bukusoglu N, Akdeniz F, Goksen D. 2004. Is obesity a risk factor
for psychopathology among adolescents? Pediatr Int. 46:296–301.
Fanselow MS, Dong HW. 2010. Are the Dorsal and Ventral Hippocampus Functionally Distinct
Structures? Neuron. 65:7–19.
Fennema-Notestine C, Gamst AC, Quinn BT, Pacheco J, Jernigan TL, Thal L, Buckner R,
Killiany R, Blacker D, Dale AM, Fischl B, Dickerson B, Gollub RL. 2007. Feasibility of
multi-site clinical structural neuroimaging studies of aging using legacy data.
Neuroinformatics. 5:235–245.
Fitzpatrick AL, Kuller LH, Lopez OL, Diehr P, O’Meara ES, Longstreth Jr. WT., Luchsinger JA.
2009. Mid- and Late-Life Obesity: Risk of Dementia in the Cardiovascular Health
Cognition Study. Arch Neurol. 66:336–342.
Frederick CB, Snellman K, Putnam RD. 2014. Increasing socioeconomic disparities in
adolescent obesity. Proc Natl Acad Sci. 111:1338–1342.
Friston KJ. 1997. Testing for Anatomically Specified Regional Effects. Hum Brain Mapp.
5:133–136.
Gasbarri A, Packard MG, Campana E, Pacitti C. 1994. Anterograde and retrograde tracing of
projections from the ventral tegmental area to the hippocampal formation in the rat. Brain
Res Bull. 33:445–452.
Gasbarri A, Verney C, Innocenzi R, Campana E, Pacitti C. 1994. Mesolimbic dopaminergic
neurons innervating the hippocampal formation in the rat: a combined retrograde tracing
and immunohistochemical study. Brain Res. 668:71–79.

124

Giedd J, Vaituzis C, Hamburger S, Lange N, Rajapakse JC, Kaysen D, Vauss YC, Rapoport JL.
1996. Quantitative MRI of the Temoral Lobe, Amydala, and Hippocampus in Normal
Human Development: Ages 4-18 Years. J Comp Neurol. 366:223–230.
Gómez RL, Edgin JO. 2016. The extended trajectory of hippocampal development: Implications
for early memory development and disorder. Dev Cogn Neurosci. 18:57–69.
Guillemot-Legris O, Muccioli GG. 2017. Obesity-induced neuroinflammation: beyond the
hypothalamus. Trends Neurosci. 40:237–253.
Hahn JD, Swanson LW. 2012. Connections of the lateral hypothalamic area juxtadorsomedial
region in the male rat. J Comp Neurol. 520:1831–1890.
Haines L, Wan KC, Lynn R, Barrett TG, Shield JPH. 2007. Rising incidence of Type 2 diabetes
in children in the U.K. Diabetes Care. 30:1097–1101.
Hao S, Dey A, Yu X, Stranahan AM. 2016. Dietary obesity reversibly induces synaptic stripping
by microglia and impairs hippocampal plasticity. Brain Behav Immun. 51:230–239.
Hargrave SL, Jones S, Davidson TL. 2016. The Outward Spiral: A vicious cycle model of
obesity and cognitive dysfunction. Curr Opin Behav Sci. 9:40–46.
Hashimoto T, Takeuchi H, Taki Y, Yokota S, Hashizume H, Asano K, Asano M, Sassa Y,
Nouchi R, Kawashima R. 2015. Increased posterior hippocampal volumes in children with
lower increase in body mass index: A 3-year longitudinal MRI study. Dev Neurosci.
37:153–160.
Haslam DW, James PT. 2005. Obesity. Lancet. 366:1197–1209.
Herting MM, Johnson C, Mills KL, Vijayakumar N, Dennison M, Liu C, Goddings AL, Dahl
RE, Sowell ER, Whittle S, Allen NB, Tamnes CK. 2018. Development of subcortical
volumes across adolescence in males and females: A multisample study of longitudinal
changes. Neuroimage. 172:194–205.
Herting MM, Nagel BJ. 2012. Aerobic fitness relates to learning on a virtual Morris Water Task
and hippocampal volume in adolescents. Behav Brain Res. 233:517–525.
Higgs S. 2002. Memory for recent eating and its influence on subsequent food intake. Appetite.
39:159–166.

125

Holland S, Schmithorst VJ, Wagner M, Lee G, Rajagopal A, Sroka M, Felicelli N, Rupert A,
Clark K, Toga A, Freund L. 2015. The C-MIND project: normative MRI and behavioral
data from children from birth to 18 years. In: Proceedings of the 21st Annumal Meeting of
the Organization for Human Brain Mapping (OHBM), Honolulu, HA.
Insausti R, Cebada-Sánchez S, Marcos P. 2010. Postnatal development of the human
hippocampal formation. Adv Anat Embryol Cell Biol. 206:1–86.
Ishikawa A, Nakamura S. 2006. Ventral hippocampal neurons project axons simultaneously to
the medial prefrontal cortex and amygdala in the rat. J Neurophysiol. 96:2134–2138.
Jacka FN, Cherbuin N, Anstey KJ, Sachdev P, Butterworth P. 2015. Western diet is associated
with a smaller hippocampus: A longitudinal investigation. BMC Med. 13:1–8.
Jean-Philippe Bastard, Mustapha Maachi, Claire Lagathu, Min Ji Kim, Martine Caron, Hubert
Vidal, Jacqueline Capeau, Feve B. 2006. Recent advances in the relationship between
obesity, inflammationa nd insulin resistance. Eur Cytokine Netw. 17:4–12.
Kanoski SE. 2012. Cognitive and neuronal systems underlying obesity. Physiol Behav. 106:337–
344.
Kanoski SE, Davidson TL. 2011. Western diet consumption and cognitive impairment: Links to
hippocampal dysfunction and obesity. Physiol Behav. 103:59–68.
Kanoski SE, Grill HJ. 2017. Hippocampus Contributions to Food Intake Control: Mnemonic,
Neuroanatomical, and Endocrine Mechanisms. Biol Psychiatry. 81:748–756.
Kanoski SE, Zhang Y, Zheng W, Davidson TL, Sciences H, Lafayette W. 2010. The Effects of a
High-Energy Diet on Hippocampal Function and Blood-Brain Barrier Integrity in the Rat.
21:207–219.
Keinath AT, Wang ME, Wann EG, Yuan RK, Dudman JT, Muzzio IA. 2014. Precise spatial
coding is preserved along the longitudinal hippocampal axis. Hippocampus. 24:1533–1548.
Kerr DS, Campbell LW, Applegate MD, Brodish A, Landfield PW. 1991. Chronic stress-induced
acceleration of electrophysiologic and morphometric biomarkers of hippocampal aging. J
Neurosci. 11:1316–1324.
Khan NA, Baym CL, Monti JM, Raine LB, Drollette ES, Scudder MR, Moore RD, Kramer AF,
Hillman CH, Cohen NJ. 2015. Central adiposity is negatively associated with hippocampal-
dependent relational memory among overweight and obese children. J Pediatr. 166:302-
308.e1.

126

Krebs NF, Himes JH, Jacobson D, Nicklas TA, Guilday P, Styne D. 2007. Assessment of child
and adolescent overweight and obesity. Pediatrics. 120 Suppl.
Lynch KM, Shi Y, Toga AW, Clark KA. 2018. Hippocampal Shape Maturation in Childhood
and Adolescence. Cereb Cortex. 1–15.
Magariños AM, McEwen BS. 1995. Stress-induced atrophy of apical dendrites of hippocampal
CA3c neurons: involvement of glucocorticoid secretion and excitatory amino acid
receptors. Neuroscience. 69:89–98.
Mayeux R, Stern Y. 2012. Epidemiology of Alzheimer disease. Cold Spring Harb Perspect Med.
2:a006239.
Mestre ZL, Eichen DM, Wierenga CE, Strong D, Boutelle KN. 2017. Hippocampal atrophy and
altered brain responses to pleasant tastes among obese compared with healthy weight
children. Int J Obes. 41:1496–1502.
Miller AA, Spencer SJ. 2014. Obesity and neuroinflammation: A pathway to cognitive
impairment. Brain Behav Immun. 42:10–21.
Nambiar S, Truby H, Davies PS, Baxter K. 2013. Use of the waist-height ratio to predict
metabolic syndrome in obese children and adolescents. J Paediatr Child Health. 49.
Nouwen A, Chambers A, Chechlacz M, Higgs S, Blissett J, Barrett TG, Allen HA. 2017.
Microstructural abnormalities in white and gray matter in obese adolescents with and
without type 2 diabetes. NeuroImage Clin. 16:43–51.
Ogden CL, Carroll MD, Kit BK, Flegal M. 2014. Prevalence of Childhood and Adult Obesity in
the United States, 2011–2012. JAMA. 311:806–814.
Ogden CL, Carroll MD, Lawman HG, Fryar CD, Kruszon-Moran D, Kit BK, Flegal KM. 2016.
Trends in Obesity Prevalence Among Children and Adolescents in the United States, 1988-
1994 Through 2013-2014. JAMA. 315:2292–2299.
Pardoe H, Pell GS, Abbott DF, Berg AT, Jackson GD. 2008. Multi-site voxel-based
morphometry: Methods and a feasibility demonstration with childhood absence epilepsy.
Neuroimage. 42:611–616.
Patenaude B, Smith SM, Kennedy DN, Jenkinson M. 2011. A Bayesian model of shape and
appearance for subcortical brain segmentation. Neuroimage. 56:907–922.
Petrovich GD, Canteras NS, Swanson LW. 2001. Combinatorial amygdalar inputs to
hippocampal domains and hypothalamic behavior systems. Brain Res Rev. 38:247–289.

127

Pinhas-Hamel O, Zeitler M. 2005. the Global Spread of Type 2 Diabetes Mellitus in. J Pediatr.
146:693–700.
Poppenk J, Evensmoen HR, Moscovitch M, Nadel L. 2013. Long-axis specialization of the
human hippocampus. Trends Cogn Sci. 17:230–240.
Rozin P, Dow S, Moscovitch M, Rajaram S. 1998. What causes humans to begin and end a
meal? A role for memory for what has been eaten , as evidenced by a study of multiple meal
eating in amnesic patients. Psychol Sci. 9:392–396.
Sack M, Lenz JN, Jakovcevski M, Biedermann S V., Falfán-Melgoza C, Deussing J, Bielohuby
M, Bidlingmaier M, Pfister F, Stalla GK, Sartorius A, Gass P, Weber-Fahr W, Fuss J, Auer
MK. 2017. Early effects of a high-caloric diet and physical exercise on brain volumetry and
behavior: a combined MRI and histology study in mice. Brain Imaging Behav. 11:1385–
1396.
Sapolsky RM. 1985. A mechanism for glucocorticoid toxicity in the hippocampus: Increased
neuronal vulnerability to metabolic insults. J Neurosci. 5:1228–1232.
Satterthwaite TD, Connolly JJ, Ruparel K, Calkins ME, Jackson C, Elliott MA, Roalf DR,
Hopsona R, Prabhakaran K, Behr M, Qiu H, Mentch FD, Chiavacci R, Sleiman PMA, Gur
RC, Hakonarson H, Gur RE. 2016. The Philadelphia Neurodevelopmental Cohort: A
publicly available resource for the study of normal and abnormal brain development in
youth. Neuroimage. 124:1115–1119.
Sekeres MJ, Winocur G, Moscovitch M. 2018. The hippocampus and related neocortical
structures in memory transformation. Neurosci Lett. Accepted m.
Shefer G, Marcus Y, Stern N. 2013. Is obesity a brain disease? Neurosci Biobehav Rev.
37:2489–2503.
Sheldon S, Levine B. 2016. The role of the hippocampus in memory and mental construction.
Ann N Y Acad Sci. 1369:76–92.
Shi Y, Lai R, Krishna S, Sicotte N, Dinov I, Toga AW. 2008. Anisotropic Laplace-Beltrami
Eigenmaps: Bridging Reed graphs and skeletons. Proc IEEE Comput Soc Conf Comput Vis
Pattern Recognit. 2008:1–7.
Shi Y, Lai R, Morra JH, Dinov I, Thompson PM, Toga AW. 2010. Robust surface reconstruction
via Laplace-Beltrami eigen-projection and boundary deformation. IEEE Trans Med
Imaging. 29:2009–2022.

128

Shi Y, Lai R, Wang DJJ, Pelletier D, Mohr D, Sicotte N, Toga AW. 2014. Metric optimization
for surface analysis in the Laplace-Beltrami embedding space. IEEE Trans Med Imaging.
33:1447–1463.
Shi Y, Morra J, Thompson P, Toga A. 2009. Inverse-consistent surface mapping with Laplace-
Beltrami Eigen-Features. Comp Imag Vis. 21:467–478.
Smith SM, Jenkinson M, Woolrich MW, Beckmann CF, Behrens TEJ, Johansen-Berg H,
Bannister PR, De Luca M, Drobnjak I, Flitney DE, Niazy RK, Saunders J, Vickers J, Zhang
Y, De Stefano N, Brady JM, Matthews PM. 2004. Advances in functional and structural
MR image analysis and implementation as FSL. Neuroimage. 23:208–219.
Spencer SJ, Tilbrook A. 2011. The glucocorticoid contribution to obesity. Stress. 14:233–246.
Stanley BG, Willett VL, Donias HW, Dee MG, Duva MA. 1996. Lateral hypothalamic NMDA
receptors and glutamate as physiological mediators of eating and weight control. Am J
Physiol - Regul Integr Comp Physiol. 270.
Strange BA, Witter MP, Lein ES, Moser EI. 2014. Functional organization of the hippocampal
longitudinal axis. Nat Rev Neurosci. 15:655–669.
Takao H, Hayashi N, Ohtomo K. 2014. Effects of study design in multi-scanner voxel-based
morphometry studies. Neuroimage. 84:133–140.
Thompson PM, Hayashi KM, De Zubicaray GI, Janke AL, Rose SE, Semple J, Hong MS,
Herman DH, Gravano D, Doddrell DM, Toga AW. 2004. Mapping hippocampal and
ventricular change in Alzheimer disease. Neuroimage. 22:1754–1766.
Van Doorn C, Macht VA, Grillo CA, Reagan LP. 2017. Leptin resistance and hippocampal
behavioral deficits. Physiol Behav. 176:207–213.
Vanderwall C, Randall Clark R, Eickhoff J, Carrel AL. 2017. BMI is a poor predictor of
adiposity in young overweight and obese children. BMC Pediatr. 17:4–9.
Voineskos AN, Winterburn JL, Felsky D, Pipitone J, Rajji TK, Mulsant BH, Chakravarty MM.
2015. Hippocampal (subfield) volume and shape in relation to cognitive performance across
the adult lifespan. Hum Brain Mapp. 36:3020–3037.
Wicklow BA, Becker A, Chateau D, Palmer K, Kozyrskij A, Sellers EAC. 2015. Comparison of
anthropometric measurements in children to predict metabolic syndrome in adolescence:
Analysis of prospective cohort data. Int J Obes. 39:1070–1078.

129

Wilkerson A, Levin ED. 1999. Ventral hippocampal dopamine D1 and D2 systems and spatial
working memory in rats. Neuroscience. 89:743–749.
Williamson LL, Bilbo SD. 2013. Chemokines and the hippocampus: A new perspective on
hippocampal plasticity and vulnerability. Brain Behav Immun. 30:186–194.
Woo C-W, Krishnan A, Wager TD. 2014. Cluster-extent based thresholding in fMRI analyses:
Pitfalls and recommendations. Neuroimage. 91:412–419.
Woolley CS, Gould E, McEwen BS. 1990. Exposure to excess glucocorticoids alters dendritic
morphology of adult hippocampal pyramidal neurons. Brain Res. 531:225–231.
Worsley KJ, Andermann M, Koulis T, MacDonald D, Evans AC. 1999. Detecting changes in
nonisotropic images. Hum Brain Mapp. 8:89–101.
Worsley KJ, Evans AC, Marrett S, Neelin P. 1992. A three-dimensional statistical analysis for
CBF activation studies in human brain. J Cereb Blood Flow Metab. 12:900–918.

130

4.7 Supplementary material

Supplementary Table 1 Acquisition parameters per site and per study
Study Site Scanner Model Scan Parameters
C-MIND Cincinnati Children's
Hospital
Philips Achieva TR=8.1ms, TE=3.7, flip angle=8,
matrix size=256x160, voxel
size=1x1x1mm, acquisition time:
5:15
PNC Children's Hospital of
Philadelphia
Siemens TrioTim TR=1810ms, TE=3.5, flip angle=9,
matrix size=256x160, voxel
size=.9375x.9375x1mm, acquisition
time:3:28
DYS University of Southern
California
Siemens matrix size=256x320, voxel
size=.7x.7x.7 mm, acquisition time:
GDM University of Southern
California
Siemens
MAGNETOM
Prisma
TR=1950 ms, TE=2.26 ms, flip
angle=90, matrix size=256x176,
voxel size=1x1x1mm, acquisition
time=4:14

131

Chapter 5
Microstructure-mesh projection: An approach for the analysis of regional hippocampal
microstructure

5.1 Abstract

The hippocampus is a complex subcortical structure critical for learning and memory and
consists of a mixture of gray and white matter. Previous shape analysis studies have shown regional
patterns of expansion and contraction during child development, however the cellular processes
that give rise to these shape changes are poorly understood. Neurite orientation dispersion and
density imaging (NODDI) is a microstructural model derived from diffusion MRI (dMRI) that is
sensitized to specific anatomical features. In the present study, we propose a novel analytical
framework that combines the regional specificity of shape analysis with the sensitivity of diffusion
models to detect differences in cytoarchitecture. The microstructure-mesh projection pipeline is a
multi-modal approach that projects intra-hippocampal dMRI parameters onto hippocampal surface
mesh vertices to (1) identify regional patterns of hippocampal microstructure and (2) enable
statistical comparison of regional hippocampal features across datasets. By combining shape
analysis with diffusion microstructure, this approach can be used to further elucidate the
contribution of cellular components to changes in hippocampal shape.

5.2 Introduction

The hippocampus is a diverse subcortical structure that consists of a mixture of gray and
white matter regions organized into cell-specific strata and functionally distinct regions, including
the cornu ammonis (CA) 1-4 subfields, dentate gyrus (DG) and subiculum (Duvernoy et al., 2013;
Insausti et al., 2010). Hippocampal regions communicate through a network of serial and recurrent
connections, including the perforant path, mossy fibers, and Schaffer collaterals (Amaral and
Insausti, 1990) and this circuitry is critical for learning, memory and other hippocampal-dependent
processes (Sheldon and Levine, 2016). The hippocampus undergoes a protracted and spatially
varying developmental trajectory, however evidence of specific cellular and cytoarchitectural

132

maturation are limited to human post-mortem and animal studies (Insausti et al., 2010; Jabès et al.,
2011; Kondo et al., 2009, 2008; Lavenex et al., 2007; Seress, 2001).
Common neuroimaging approaches to study human hippocampal structural development
in vivo rely on the detection of macroscopic changes to the size of the whole hippocampus or
subfields using T1-weighted (T1w) tissue contrasts (DeMaster et al., 2014; Gogtay et al., 2006;
Lin et al., 2013; Lynch et al., 2018; Uematsu et al., 2012). Previous studies on the relationship
between whole hippocampal volume and age across child development yield mixed results
(DeMaster et al., 2014; Giedd et al., 1996; Knickmeyer and Gouttard, 2008; Uematsu et al., 2012;
Yurgelun-Todd et al., 2003). It is therefore likely that relevant developmental processes within the
hippocampus are obscured when considered as a volumetric unit because the hippocampus is a
heterogeneous structure that does not mature uniformly. The limitations of volumetry are
overcome by morphological approaches, such as shape analysis, due to the ability to detect
regional changes in structure. Shape analysis is an approach that maps intrinsic geometric features
on surface representations derived from segmented brain structures (Shi et al., 2014) and using
this approach, we previously demonstrated regional patterns of hippocampal surface expansion
across childhood. However, shape analysis is inherently non-specific, and a given change in
hippocampal shape may reflect a number of structural phenomena within the underlying tissue.
Diffusion MRI (dMRI) relies on water diffusion patterns to derive indirect measures of
microstructural features and is commonly used to elucidate white matter structure (Beaulieu,
2002). Recent advances in diffusion MRI have led to the development of multi-shell
microstructural methods that model tissue compartments, such as neurite orientation dispersion
and density imaging (NODDI) (Zhang et al., 2012). NODDI generates two parameters sensitized
to distinct phenomena: the neurite density index (NDI) estimates the fraction of tissue that
comprises intracellular compartments and reflects neurite density, while the orientation dispersion
(ODI) measures the degree of angular variation and provides an index of geometric complexity
(Zhang et al., 2012). The utility of NODDI extends beyond white matter applications and previous
studies have assessed different aspects of cortical and subcortical microstructure using NODDI
(Fukutomi et al., 2018; Mah et al., 2017; Nazeri et al., 2015).Because NODDI models the full
spectrum of neurite orientations (Zhang et al., 2012), it can therefore represent the dispersed
organization of cell bodies and dendrites that occupy gray matter.

133

Here, we present a framework to combine the localized specificity of shape analysis with
the microstructural sensitivity obtained with diffusion MRI (dMRI) models to enable researchers
to: (1) visualize regional microstructural patterns on the surface of the hippocampus and (2)
analyze changes in local hippocampal microstructure. DTI and NODDI parameters are projected
onto a surface representation of the hippocampus and are anatomically aligned to a population-
based template for one-to-one correspondence using Metric Optimization for Computational
Analysis (MOCA; https://www.nitrc.org/projects/moca_2015/) . By combining shape analysis
with diffusion microstructure, we demonstrate the ability to identify local microstructural
signatures of hippocampal regions.

5.3 Methods

Subjects

Cross-sectional neuroimaging data were obtained through the publicly available Human
Connectome Project (HCP) consortium led by Washington University, University of Minnesota
and Oxford University (Van Essen et al., 2013). Scans from 174 typical young adults were used
for the present analysis (91 female, range: 22-36 years, M=28.6, SD=3.8 years).

MRI acquisition

Participants were scanned on a customized Siemens 3T Connectome Skyra at Washington
University using a standard 32-channel Siemens receive head coil (Van Essen et al., 2013).
Participants completed two T1-weighted scans utilizing a 3D MPRAGE sequence with 0.7 mm
isotropic resolution (FOV = 224 mm, matrix = 320x256, TR = 2400 ms, TE = 2.14 ms, TI = 1000
ms, FA = 8°) (Glasser et al., 2013). Multi-shell DWI scans were acquired using a multi-band
acquisition and a spin-echo Stejskal–Tanner sequence with two b-shells of about 1000 and 2500
s/mm
2
with 76 and 75 gradient directions, respectively (TR=3.67 s , TE=74.8 ms, matrix=140x120,
voxel resolution=1.5 mm isotropic)

134

Data preprocessing

Quality assurance procedures, motion correction, and intensity normalization were
performed on imaging data using the HCP’s Minimal Processing Pipeline (Glasser et al., 2013).
All brain volumes were skull-stripped using FSL’s BET (Battaglini et al., 2008). For each subject,
the two DWIs were co-registered to a common space with corrected gradient tables using AIR
(Woods et al., 1998). For each participant, native DWI were aligned to native T1w images using
boundary-based registration (BBR) with FSL’s FLIRT (Greve and Fischl, 2009). This approach
utilizes reliable T1w intensity contrasts in white matter boundaries to improve the accuracy of
multi-modal registration. Bilateral hippocampal volumes were segmented from the T1w volumes
with FSL FIRST (Patenaude et al., 2011) and were then binarized to create volumetric masks of
the hippocampus.

dMRI parameters

The axonal water fraction (AWF) and orientation dispersion index (ODI) were estimated
using in-house software based on the NODDI Matlab Toolbox (Zhang et al., 2012). The DTI
parameters fractional anisotropy (FA), axial diffusivity (AD), radial diffusivity (RD) and mean
diffusivity (MD) were computed using a two-stage weighted-least squares estimation scheme
(Veraart et al., 2013). DTI and NODDI parameter maps were then aligned to the native T1w space
by applying the BBR transform used for the DWI.

Microstructure-mesh generation

Shape analysis was performed using Metric Optimization for Computational Anatomy
(MOCA) software (Shi et al., 2014), which utilizes Laplace-Beltrami eigen-functions as isometry-
invariant descriptors of intrinsic surface geometry. Hippocampal shape can be described using its
signed distance function (SDF) (Bertalmío et al., 2001; Memoli et al., 2004a, 2004b; Osher and
Sethian, 1988), where the zero level set with genus zero topology represents the boundary of the
hippocampus (Shi et al., 2007). The hippocampal surface is then represented with a triangulated
mesh through an iterative process of outlier detection, surface deformation, and reassignment of

135

vertex location (Shi et al., 2010). All hippocampal meshes were resampled to 2000 vertices, and
were averaged together in a common space to generate population-average templates for each
hemisphere using SurfStat implemented in Matlab (www.math.mcgill.ca/keith/surfstat).

Figure 5.1 Partial volume issues with sampling microstructure at the zero level set surface.
(Left) Coronal slice of an FA map from a representative subjects. Bilateral hippocampal surfaces
are delineated in green. (Right) Zoomed in view of the left hippocampus delineated in green. The
yellow arrow shows an example region that shows partial voluming of external white matter in the
hippocampus.

The hippocampus borders ventricles, white matter tracts, cortex, and subcortical structures
throughout its length (Insausti et al., 2010). Therefore, sampling dMRI microstructural parameters
at the level of the mesh will risk partial voluming from extra-hippocampal tissue (Figure 5.1). To
overcome this limitation, we propose sampling dMRI values from a level set (iso-surface) of the
original hippocampal surface. The SDF defines a surface as a level set of a volumetric function, ϕ
: U ⟼ ℝ and for a closed iso-surface Sk, is:
Sk = {s| ϕ(s)=k} (1)

where k is the iso-value defining the level set (Sethian, 1999). Level sets from SDFs provide the
mathematical mechanism for computing surface deformations according to specified iso-values of
ϕ. In order words, the shortest distance from a given point on the iso-surface Sk to the zero level
set surface, S0, will be constant and equal to k. Furthermore, the sign of k determines whether the
iso-surface undergoes outward shape deformation (positive) or inward shape deformation
(negative). Therefore for the purposes of feature sampling inside the hippocampus, the iso-surface

136

with k=-2 was computed to generate a triangulated mesh within the hippocampal surface using
MOCA (Shi et al., 2014) (Figure 5.2). The atlas mesh was then projected onto each subject’s iso-
surface using conformal maps to enable correspondence between subjects.
For NODDI and DTI parameter maps aligned to native T1w space, the gridded data was
interpolated on the iso-surface vertices using linear interpolation in Matlab. All pullback function
is them applied to project the interpolated dMRI microstructural features from the iso-surface to
the population template surface vertices for statistical analysis (Shi et al., 2014).

Figure 5.2 Generation of inner iso-surface for parameter interpolation.
(A, left) T1w scans with hippocampal surfaces delineated in green and (A, right) SDF maps color-
coded by level set (k) with the zero level set surface outlined in black. SDF maps are shown for a
representative participant in (A, top) a coronal slice and (A, bottom) a sagittal slice. (B)
Comparison of triangulated meshes generated from the zero level set in green and the inner iso-
surface for k=-2 in red.

Statistical analysis

NODDI and DTI features on the population template were averaged across vertices to
generate mean hippocampal microstructure maps. To explore developmental changes in vertex-
wise feature changes, general linear models were applied to each vertex to test for the main effect
of age while controlling for sex. Random field theory (RFT) was used to identify the spatial extent
of significant clusters (Cao and Worsley, 1999; Worsley et al., 1999). A supra-threshold cluster
level of p<.001 and a set level threshold of p<.05 were used to classify the height and spatial extent
of the cluster.

137

5.4 Results

Average dMRI feature maps demonstrate spatial variability in the distribution of DTI and
NODDI parameters (Figure 5.3). FA, NDI and ODI show different spatial distributions on the
hippocampus. The hippocampus shows low FA throughout the majority of the hippocampus
(FA=.1), however elevated FA is observed on the superior side of the hippocampal tail (Figure
5.3a). Mean NDI is low and uniform on the inferior side of the hippocampus, with NDI elevated
in the superior and mesial hippocampus (Figure 5.3b). ODI was lowest in the lateral superior
hippocampus and highest in the head and along the inferior hippocampus (Figure 5.3c).

Figure 5.3 Hippocampal microstructure maps from the microstructure-mesh projection
approach
Distribution of dMRI microstructural features on the hippocampal surface averaged across all
participants for (A) DTI FA, (B) NODDI AWF and (C) NODDI ODI

5.5 Discussion

In this study, we developed a novel analytical framework for hippocampal surface mapping
with quantitative dMRI features. This approach combines the regional specificity of morphological
surface mapping with the microstructural sensitivity afforded with diffusion models, such as DTI
and NODDI. Our results show regional variability in the distribution of dMRI features across
bilateral hippocampi. Furthermore, our results create a framework for the statistical analysis of the
spatial distribution of regional dMRI microstructure in the hippocampus.
Across subjects, mean FA was highest in the superior posterior regions of the hippocampus
that likely reflect putative CA1 and CA2-3 subfields (Duvernoy et al., 2013). Selectively elevated

138

FA in this region may reflect the myelinated intra-hippocampal white matter paths of the Schaffer
collaterals, which project from CA3 neurons to the apical dendrites of CA1 through the stratum
radiatum of CA1-3 (Insausti et al., 2010; Kondo et al., 2009). The Schaffer collaterals make up a
relay in the polysynaptic pathway that consists of: (1) perfornat pathway projections from the
entorhinal cortex (ERC) to the DG, (2) mossy fiber projections from the DG to CA3, and (3)
Schaffer collaterals (Amaral and Insausti, 1990; Duvernoy et al., 2013). FA is sensitive to tissue
features that alter the degree of anisotropy, and increased FA can be due to an increase in axonal
coherence, myelination, density, or a combination of tissue features (Beaulieu, 2009). Unlike the
complex geometry of unmyelinated mossy fiber (Seress and Ribak, 1995), the Schaffer collaterals
have dense, highly organized projections. In a previous study using polarized light imaging to
visualize hippocampal fibers in post-mortem human hippocampal tissue samples, Schaffer
collaterals were described as a dense band extending from CA3 to CA1 (Zeineh et al., 2017). In
addition to the Schaffer collaterals, increased FA in the superior tail may reflect myelinated CA3
outputs through the fornix (Saunders and Aggleton, 2007).
NDI is a specific measure of the fraction of tissue that comprises axons and dendrites and
reflects an index of neurite density (Sepehrband et al., 2015; Zhang et al., 2012). Elevated NDI
was observed in the superior body on the mesial side of the hippocampus, which likely consist of
putative CA2-3 subfields and the DG (Duvernoy et al., 2013). The CA2 subfield is characterized
by larger pyramidal cells and increased cell packing density relative to neighboring subfields
(Insausti et al., 2010), which would occupy a higher fraction of the intracellular space modeled by
NDI. Furthermore, adult neurogenesis occurs in the subgranular zone (SGZ) immediately adjacent
to the granule cell layer in the DG and this may lead to an increase in neurite density (Eriksson et
al., 1998). Adult neural stem cells in the SGZ generate proliferating intermediate progenitor cells
that can undergo morphogenesis and differentiation into mature DG neurons (Gonçalves et al.,
2016). Granule cells in the DG are tightly packed and characterized by their small elliptical cell
bodies, tree of apical dendrites that project into the DG stratum moleculare, and unmyelinated
axons with thorny excrescences that form the mossy fiber projects to CA3 (Seress and Ribak,
1995). It is possible that NDI shows more spread and variability compared to FA because NDI is
sensitive to processes obscured with FA. FA is sensitive to the general displacement of water,
however a single tensor is not sufficient to model crossing fibers (Beaulieu, 2009; Jones et al.,
2013). On the other hand, NODDI models the full spectrum of orientation dispersion by modeling

139

the intracellular compartment as a set of zero-radius sticks (Zhang et al., 2012). Therefore, it is
possible that NDI is more sensitive to complex fiber projections, such as mossy fibers, compared
to FA.
ODI quantifies the angular variation of neurite orientation using a Watson distribution and
ranges from strictly parallel orientations (ODI=0) to isotropically-dispersed orientations (ODI=1)
(Zhang et al., 2012). In the hippocampus, ODI demonstrated an approximately inverted pattern
compared to FA and NDI, with high ODI observed in the inferior hippocampus and head and low
ODI in the superior tail. Decreased ODI in the tail occupies the same location as elevated FA, and
likely reflects the coherent myelinated bundles of the Schaffer collaterals (Duvernoy et al., 2013).
The increased ODI observed in the hippocampal head corresponds to CA1-4, DG and subiculum,
however the subiculum accounts for the single largest cytoarchitectonic field in the head (Insausti
and Amaral, 2004). Furthermore, the head consists of variable macroscopic digitations, where
hippocampal structures bend and flex into complex orientations (Gertz, 1972; Insausti et al., 2010).
High orientation dispersion in the inferior hippocampus likely reflects predominantly the
subiculum and some CA1 subfield. The subiculum is composed of three layers, including a deep
polymorphic layer, a large pyramidal cell layer, and a molecular layer that is continuous with the
CA1 stratum lacunosum (Duvernoy et al., 2013). The principle cells in the pyramidal layer are
more loosely packed and disorganized compared to the CA1 subfield (Insausti et al., 2010), which
may contribute to environment of isotropy observed with high ODI. The hippocampus is
innervated by the ERC in the inferior hippocampus where the perforant pathway projects to the
DG by perforating through the subiculum and interrupting the subicular architectural organization
(Amaral and Insausti, 1990). Furthermore, CA1 have extensive radial inputs to the subiculum that
are organized perpendicular to the perforant path (Honda and Shibata, 2017), which could
contribute to elevated ODI in this region.
Together, these results suggest that dMRI features sensitized to different cytoarchitectural
phenomena demonstrate different spatial patterns of hippocampal microstructure. FA may be
uniquely sensitive to the coherently organized bundles of the Schaffer collaterals, while NDI can
detect regions with elevated neurite density, irrespective of local orientation distributions,
including the mossy fibers and processes associated with SGZ neurogenesis in the DG. ODI shows
the spatial organization of hippocampal architecture, and may therefore describe regions with
macroscopic and microscopic orientational complexity, such as the hippocampal head and

140

subiculum, respectively. Furthermore, the microstructure-mesh projection pipeline is not restricted
to NODDI and DTI; any quantitative MRI parameter and contrast that represents a biophysical
model can be projected into the hippocampal surface. Magnetization transfer (MT), an MRI
contrast sensitized to hydrogen atoms bound to macromolecules, provides an index of myelin
concentration (Schmierer et al., 2004) and previous studies have demonstrated its sensitivity to
cortical myelination (Dick et al., 2012; Stüber et al., 2014). This contrast can be used to elucidate
global myelination patterns within the hippocampus. Additionally, MT parameters can be
combined with NDI from NODDI to generate a two-compartment sensitized to the g-ratio (Stikov
et al., 2015) or the ratio of axon diameter to myelin thickness, that is directly related to axonal
function and conduction speeds (Rushton, 1951). Susceptibility-weighted imaging (SWI) provides
a direct measure of non-haem iron (Duyn, 2013; Reichenbach, 2012), which plays an important
role in brain function and can be indicative of degenerative processes (Deistung et al., 2013;
Todorich et al., 2009). Evidence from human post-mortem tissue shows iron depositions and
neurodegeneration mediated by microglia within the AD hippocampus (Zeineh et al., 2015), and
the microstructure-mesh projection approach can be used to visualize the spatial distribution of
iron deposition by microglia to identify vulnerable structures.
While the present study provides a novel framework to visualize regional patterns of
hippocampal microstructure, future studies should determine how these parameters change across
child development. Shape analysis demonstrates dynamic and nonlinear morphological changes
on the hippocampal surface across development (Lynch et al., 2018), however it is unclear what
cellular processes give rise to the observed surface deformations. Age-related changes in
biophysical models sensitized to cell density, myelin content, and cytoarchitecture (Weiskopf et
al., 2015) on regional hippocampal structure can be used to determine which model best explains
regional hippocampal expansion observed in childhood and adolescence (Lynch et al., 2018).
In conclusion, the present study demonstrates the feasibility and utility of the
microstructure-mesh projection pipeline to explore regional patterns of hippocampal
cytoarchitecture using dMRI models and shape analysis. By sampling quantitative dMRI
parameters in an inner iso-surface generated with hippocampal level sets (Sethian, 1999; Shi et al.,
2014), this approach ensures the microstructure projection reflects cellular features within the
hippocampus and avoids partial volume effects. Our results demonstrate regional variability in FA,
NDI and ODI distributions that reflect the known organization and distinct cellular features

141

observed in hippocampal subfields. This approach is ideal for the in vivo study of human
hippocampal features with sub-voxel resolution that are often obscured with clinically available
acquisitions. These results contribute to our understanding of the spatial distribution of
hippocampal processes and the proposed technique provides regional specificity for the detection
of microstructural differences related to development, function, and neurological disease.

5.6 References

Amaral, D., Insausti, R., 1990. Hippocampal formation, in: Praxinos, G. (Ed.), The Human
Nervous System. Academic Press, San Diego, pp. 711–755.
Battaglini, M., Smith, S.M., Brogi, S., De Stefano, N., 2008. Enhanced brain extraction improves
the accuracy of brain atrophy estimation. Neuroimage 40, 583–589.
https://doi.org/10.1016/j.neuroimage.2007.10.067
Beaulieu, C., 2009. The biological basis of diffusion anisotropy, in: In: Johansen-Berg, H.,
Behrens, T.E.J. (Eds. . (Ed.), Diffusion MRI: From Quantitative Measurement to In-Vivo
Neuroanatomy. Academic Press, pp. 105–129. https://doi.org/10.1016/B978-0-12-396460-
1.00008-1
Beaulieu, C., 2002. The basis of anisotropic water diffusion in the nervous system - A technical
review. NMR Biomed. 15, 435–455. https://doi.org/10.1002/nbm.782
Bertalmío, M., Cheng, L., Osher, S., Sapiro, G., 2001. Variational problems and partial
differential equations on implicit surfaces. J. Comput. Phys. 174, 759–780.
Cao, J., Worsley, K.J., 1999. The detection of local shape changes via the geometry of
Hotelling’s T2 fields. Ann. Stat. 27, 925–942.
Deistung, A., Schweser, F., Wiestler, B., Abello, M., Roethke, M., Sahm, F., Wick, W., Nagel,
A.M., Heiland, S., Schlemmer, H.P., Bendszus, M., Reichenbach, J.R., Radbruch, A., 2013.
Quantitative Susceptibility Mapping Differentiates between Blood Depositions and
Calcifications in Patients with Glioblastoma. PLoS One 8, 1–8.
https://doi.org/10.1371/journal.pone.0057924
DeMaster, D., Pathman, T., Lee, J.K., Ghetti, S., 2014. Structural development of the
hippocampus and episodic memory: Developmental differences along the anterior/posterior
axis. Cereb. Cortex 24, 3036–3045. https://doi.org/10.1093/cercor/bht160

142

Dick, F., Tierney, A.T., Lutti, A., Josephs, O., Sereno, M.I., Weiskopf, N., 2012. In vivo
functional and myeloarchitectonic mapping of human primary auditory areas. J. Neurosci.
32, 16095–16105. https://doi.org/10.1523/JNEUROSCI.1712-12.2012.In
Duvernoy, H., Cattin, F., Risold, P.-Y., 2013. The Human Hippocampus: Functional anatomy,
vascularization and serial sections with MRI, 4th Ed. ed. Springer, Berlin, Germany.
Duyn, J., 2013. MR susceptibility imaging. J. Magn. Reson. 229, 198–207.
https://doi.org/10.1126/scisignal.2001449.Engineering
Eriksson, P., Perfilieva, E., Bjork-Eriksson, T., Aborn, A., Nordborg, C., Peterson, D., Gage, F.,
1998. Neurogenesis in the adult human hippocampus. Nat. Med. 4, 1313–1317.
Fukutomi, H., Glasser, M.F., Zhang, H., Autio, J.A., Coalson, T.S., Okada, T., Togashi, K., Van
Essen, D.C., Hayashi, T., 2018. Neurite imaging reveals microstructural variations in
human cerebral cortical gray matter. Neuroimage 1–12.
https://doi.org/10.1016/j.neuroimage.2018.02.017
Gertz, S., 1972. Structural variations in the rostral human hippocampus. Johns Hopkins Med. J.
130, 367–376.
Giedd, J., Vaituzis, C., Hamburger, S., Lange, N., Rajapakse, J.C., Kaysen, D., Vauss, Y.C.,
Rapoport, J.L., 1996. Quantitative MRI of the Temoral Lobe, Amydala, and Hippocampus
in Normal Human Development: Ages 4-18 Years. J. Comp. Neurol. 366, 223–230.
https://doi.org/10.1002/(SICI)1096-9861(19960304)366
Glasser, M.F., Sotiropoulos, S.N., Wilson, J.A., Coalson, T.S., Fischl, B., Andersson, J.L., Xu,
J., Jbabdi, S., Webster, M., Polimeni, J.R., Van Essen, D.C., Jenkinson, M., 2013. The
minimal preprocessing pipelines for the Human Connectome Project. Neuroimage 80, 105–
124. https://doi.org/10.1016/j.neuroimage.2013.04.127
Gogtay, N., Nugent, T.F., Herman, D.H., Ordonez, A., Greenstein, D., Hyashi, K.M., Clasen, L.,
Toga, A.W., Giedd, J.N., Rapoport, J.L., Thompson, P.M., 2006. Dynamic Mapping of
Normal Human Hippocampal Development. Hippocampus 16, 664–72.
https://doi.org/10.1002/hipo
Gonçalves, J.T., Schafer, S.T., Gage, F.H., 2016. Adult Neurogenesis in the Hippocampus: From
Stem Cells to Behavior. Cell 167, 897–914. https://doi.org/10.1016/j.cell.2016.10.021
Greve, D.N., Fischl, B., 2009. Accurate and robust brain image alignment using boundary-based
registration. Neuroimage 48, 63–72. https://doi.org/10.1016/j.neuroimage.2009.06.060

143

Honda, Y., Shibata, H., 2017. Organizational connectivity among the CA1, subiculum,
presubiculum, and entorhinal cortex in the rabbit. J Comp Neurol 525, 3705–3741.
Insausti, R., Amaral, D., 2004. Hippocampal formation, in: The Human Nervous System. pp.
871–906.
Insausti, R., Cebada-Sánchez, S., Marcos, P., 2010. Postnatal development of the human
hippocampal formation. Adv. Anat. Embryol. Cell Biol. 206, 1–86.
Jabès, A., Lavenex, P.B., Amaral, D.G., Lavenex, P., 2011. Postnatal development of the
hippocampal formation: A stereological study in macaque monkeys. J. Comp. Neurol. 519,
1051–1070. https://doi.org/10.1002/cne.22549
Jones, D.K., Knösche, T.R., Turner, R., 2013. White matter integrity, fiber count, and other
fallacies: the do’s and don’ts of diffusion MRI. Neuroimage 73, 239–54.
https://doi.org/10.1016/j.neuroimage.2012.06.081
Knickmeyer, R., Gouttard, S., 2008. A structural MRI study of human brain development from
birth to 2 years. J. Neurosci. 28, 12176–12182. https://doi.org/10.1523/JNEUROSCI.3479-
08.2008.A
Kondo, H., Lavenex, P., Amaral, D.G., 2009. Intrinsic connections of the macaque monkey
hippocampal formation: II. CA3 connections. J Comp Neurol 515, 349–377.
https://doi.org/10.1002/cne.22056.Intrinsic
Kondo, H., Lavenex, P., Amaral, D.G., 2008. Intrinsic connections of the macaque monkey
hippocampal formation: I. Dentate gyrus. J Comp Neurol 511, 497–520.
https://doi.org/10.1002/cne.21825.Intrinsic
Lavenex, P., Banta Lavenex, P., Amaral, D.G., 2007. Postnatal development of the primate
hippocampal formation. Dev. Neurosci. 29, 179–192. https://doi.org/10.1159/000096222
Lin, M., Fwu, P.T., Buss, C., Davis, E.P., Head, K., Muftuler, L.T., Sandman, C.A., Su, M.-Y.,
2013. Developmental changes in hippocampal shape among preadolescent children. Int. J.
Dev. Neurosci. 31, 473–81. https://doi.org/10.1016/j.ijdevneu.2013.06.001
Lynch, K.M., Shi, Y., Toga, A.W., Clark, K.A., 2018. Hippocampal Shape Maturation in
Childhood and Adolescence. Cereb. Cortex 1–15. https://doi.org/10.1093/cercor/bhy244
Mah, A., Geeraert, B., Lebel, C., 2017. Detailing neuroanatomical development in late childhood
and early adolescence using NODDI. PLoS One 12, 1–16.
https://doi.org/10.1371/journal.pone.0182340

144

Memoli, F., Sapiro, G., Osher, S., 2004a. Solving variational problems and partial differential
equations mapping into general target manifolds. ournal Comput. Phys. 195, 263–292.
Memoli, F., Sapiro, G., Thompson, P., 2004b. Implicit brain imaging. Neuroimage 23, S179–
S188.
Nazeri, a., Chakravarty, M.M., Rotenberg, D.J., Rajji, T.K., Rathi, Y., Michailovich, O. V.,
Voineskos, a. N., 2015. Functional Consequences of Neurite Orientation Dispersion and
Density in Humans across the Adult Lifespan. J. Neurosci. 35, 1753–1762.
https://doi.org/10.1523/JNEUROSCI.3979-14.2015
Osher, S., Sethian, J., 1988. Fronts propagation with curvature-dependent speed: algorithms
based on Hamilton-Jacobi formulations. J. Comput. Phys. 79, 12–49.
Patenaude, B., Smith, S.M., Kennedy, D.N., Jenkinson, M., 2011. A Bayesian model of shape
and appearance for subcortical brain segmentation. Neuroimage 56, 907–922.
https://doi.org/10.1016/j.neuroimage.2011.02.046
Reichenbach, J.R., 2012. The future of susceptibility contrast for assessment of anatomy and
function. Neuroimage 62, 1311–1315. https://doi.org/10.1016/j.neuroimage.2012.01.004
Rushton, A.H., 1951. A theory of the effects of fibre size in medullated nerve. J Physiol 115,
101–122.
Saunders, R., Aggleton, J., 2007. Origin and topography of fibers contributing to the fornix in
macaque monkeys. Hippocampus 17, 396–411.
Schmierer, K., Scaravilli, F., Altmann, D.R., Barker, G.J., Miller, D.H., 2004. Magnetization
transfer ratio and myelin in postmortem multiple sclerosis brain. Ann. Neurol. 56, 407–415.
https://doi.org/10.1002/ana.20202
Sepehrband, F., Clark, K. a., Ullmann, J.F.P., Kurniawan, N.D., Leanage, G., Reutens, D.C.,
Yang, Z., 2015. Brain tissue compartment density estimated using diffusion-weighted MRI
yields tissue parameters consistent with histology. Hum. Brain Mapp. 00, n/a-n/a.
https://doi.org/10.1002/hbm.22872
Seress, L., 2001. Morphological changes of the human hippocampal formation from
midgestation to early childhood, in: Nelson, A.A., Luciana, M. (Eds.), Handbook of
Developmental Cognitive Neuroscience. MIT Press, Cambridge, MA, pp. 45–47.
Seress, L., Ribak, C.E., 1995. Postnatal-Development of CA3 Pyramidal Neurons and Their
Afferents in the Ammons Horn of Rhesus-Monkeys. Hippocampus 5, 217–31.

145

Sethian, J., 1999. Level Set Methods and Fast Marching Methods. Cambridge University Press,
Cambridge, UK.
Sheldon, S., Levine, B., 2016. The role of the hippocampus in memory and mental construction.
Ann. N. Y. Acad. Sci. 1369, 76–92. https://doi.org/10.1111/nyas.13006
Shi, Y., Lai, R., Morra, J.H., Dinov, I., Thompson, P.M., Toga, A.W., 2010. Robust surface
reconstruction via Laplace-Beltrami eigen-projection and boundary deformation. IEEE
Trans. Med. Imaging 29, 2009–22. https://doi.org/10.1109/TMI.2010.2057441
Shi, Y., Lai, R., Wang, D.J.J., Pelletier, D., Mohr, D., Sicotte, N., Toga, A.W., 2014. Metric
optimization for surface analysis in the Laplace-Beltrami embedding space. IEEE Trans.
Med. Imaging 33, 1447–63. https://doi.org/10.1109/TMI.2014.2313812
Shi, Y., Thompson, P.M., de Zubicaray, G.I., Rose, S.E., Tu, Z., Dinov, I., Toga, A.W., 2007.
Direct mapping of hippocampal surfaces with intrinsic shape context. Neuroimage 37, 792–
807. https://doi.org/10.1016/j.neuroimage.2007.05.016
Stikov, N., Campbell, J.S.W., Stroh, T., Lavelée, M., Frey, S., Novek, J., Nuara, S., Ho, M.-K.,
Bedell, B.J., Dougherty, R.F., Leppert, I.R., Boudreau, M., Narayanan, S., Duval, T.,
Cohen-Adad, J., Picard, P., Gasecka, A., Côté, D., Bruce Pike, G., 2015. In vivo histology
of the myelin g-ratio with magnetic resonance imaging. Neuroimage.
https://doi.org/10.1016/j.neuroimage.2015.05.023
Stüber, C., Morawski, M., Schäfer, A., Labadie, C., Wähnert, M., Leuze, C., Streicher, M.,
Barapatre, N., Reimann, K., Geyer, S., Spemann, D., Turner, R., 2014. Myelin and iron
concentration in the human brain: A quantitative study of MRI contrast. Neuroimage 93,
95–106. https://doi.org/10.1016/j.neuroimage.2014.02.026
Todorich, B., Pasquini, J.M., Garcia, C.I., Paez, P.M., Connor, J.R., 2009. Oligodendrocytes and
myelination: The role of iron. Glia 57, 467–478. https://doi.org/10.1002/glia.20784
Uematsu, A., Matsui, M., Tanaka, C., Takahashi, T., Noguchi, K., Suzuki, M., Nishijo, H., 2012.
Developmental Trajectories of Amygdala and Hippocampus from Infancy to Early
Adulthood in Healthy Individuals. PLoS One 7, e46970.
https://doi.org/10.1371/journal.pone.0046970
Van Essen, D.C., Smith, S.M., Barch, D.M., Behrens, T.E.J., Yacoub, E., Ugurbil, K., 2013. The
WU-Minn Human Connectome Project: An overview. Neuroimage 80, 62–79.
https://doi.org/10.1016/j.neuroimage.2013.05.041

146

Veraart, J., Sijbers, J., Sunaert, S., Leemans, A., Jeurissen, B., 2013. Weighted linear least
squares estimation of diffusion MRI parameters : Strengths , limitations , and pitfalls.
Neuroimage 81, 335–346. https://doi.org/10.1016/j.neuroimage.2013.05.028
Weiskopf, N., Mohammadi, S., Lutti, A., Callaghan, M.F., 2015. Advances in MRI-based
computational neuroanatomy: from morphometry to in-vivo histology. Curr. Opin. Neurol.
28, 313–22. https://doi.org/10.1097/WCO.0000000000000222
Woods, R.P., Grafton, S.T., Holmes, C.J., Cherry, S.R., Mazziotta, J.C., 1998. Automated image
registration: I. General methods and intrasubject, intramodality validation. J. Comput.
Assist. Tomogr. 22, 139–152.
Worsley, K.J., Andermann, M., Koulis, T., MacDonald, D., Evans, A.C., 1999. Detecting
changes in nonisotropic images. Hum. Brain Mapp. 8, 89–101.
Yurgelun-Todd, D.A., Killgore, W.D.S., Cintron, C.B., 2003. Cognitive correlates of medial
temporal lobe development across adolescence: A magnetic resonance imaging study.
Percept. Mot. Skills 96, 3–17.
Zeineh, M., Chen, Y., Kitzler, H., Hammond, R., Vogel, H., Rutt, B., 2015. Activated iron-
containing microglia in the human hippocampus identified by magnetic resonance imaging
in Alzheimer disease. Neurobiol Aging 36, 2483–2500.
Zeineh, M.M., Palomero-Gallagher, N., Axer, M., Gräβel, D., Goubran, M., Wree, A., Woods,
R., Amunts, K., Zilles, K., 2017. Direct Visualization and Mapping of the Spatial Course of
Fiber Tracts at Microscopic Resolution in the Human Hippocampus. Cereb. Cortex 27,
1779–1794. https://doi.org/10.1093/cercor/bhw010
Zhang, H., Schneider, T., Wheeler-Kingshott, C.A., Alexander, D.C., 2012. NODDI: practical in
vivo neurite orientation dispersion and density imaging of the human brain. Neuroimage 61,
1000–16. https://doi.org/10.1016/j.neuroimage.2012.03.072

Abstract (if available)

Abstract The goal of this dissertation is to characterize regional developmental patterns of white matter and hippocampal structure from childhood through adolescence using in vivo MRI methods sensitized to specific cellular features. The magnitude and timing of major white matter tract maturation using NODDI is described from infancy through adolescence (Chapter 2). White matter fibers undergo protracted NDI maturation and NDI developmental rates vary along the lengths of individual tracts. Next, hippocampal development is characterized from infancy through adolescence in a large multi-study cohort using shape analysis (Chapter 3). Hippocampal shape undergoes nonlinear and regionally specific patterns of surface expansion and demonstrates localized sex differences. Localized patterns of hippocampal atrophy were also associated with adiposity measures in typically developing children, demonstrating potential hippocampal structural vulnerabilities in response to childhood obesity (Chapter 4). Shape analysis is inherently non-specific, however, and a given change in hippocampal shape may reflect a number of structural phenomena within the underlying tissue. Therefore, in order to elucidate and better understand the cellular features that give rise to macroscopic changes to hippocampal size, I propose a novel analytical framework that combines the regional sensitivity afforded with shape analysis and the microstructural sensitivity of diffusion models (Chapter 5). Together, these experiments demonstrate the dynamic structural development of hippocampal and white matter processes during childhood and adolescence and offer a methodological approach to more precisely map microstructural features that reflect biological properties.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Unveiling the white matter microstructure in 22q11.2 deletion syndrome with diffusion magnetic resonance imaging

PDF

Neural & immune consequences of prenatal & childhood air pollution exposure: individual factors of risk & resilience in adolescence

PDF

Shape, pose, and connectivity in subcortical networks across the human lifespan

PDF

Diffusion MRI of the human brain: signal modeling and quantitative analysis

PDF

Pathomechanics of femoroacetabular impingement syndrome: utilizing subject-specific modeling approaches to investigate the influence of hip joint morphology and neuromuscular control

PDF

Characterization of the brain in early childhood

PDF

Simulating electrical stimulation and recording in a multi-scale model of the hippocampus

PDF

Sleep and weight gain in adolescence

PDF

Characterizing the hippocampal synaptic and sleep abnormalities of a mouse model of human chromosome 16p11.2 microdeletion

PDF

3D vessel mapping techniques for retina and brain as an early imaging biomarker for small vessel diseases

PDF

Exploring the molecular and cellular underpinnings of organ polarization using feather as the model system

Asset Metadata

Creator Lynch, Kirsten Mary (author)

Core Title Morphological and microstructural models of typical development in the hippocampus and white matter

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Neuroscience

Publication Date 12/12/2019

Defense Date 09/10/2019

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag brain morphology,Child development,diffusion MRI,hippocampus,microstructural models,OAI-PMH Harvest,shape analysis,structural MRI,white matter

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Toga, Arthur (committee chair), Clark, Kristi (committee member), Levitt, Pat (committee member), Shi, Yonggang (committee member)

Creator Email kirsten.lynch@loni.usc.edu,kirsten.m.lynch@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-250436

Unique identifier UC11673364

Identifier etd-LynchKirst-8057.pdf (filename),usctheses-c89-250436 (legacy record id)

Legacy Identifier etd-LynchKirst-8057.pdf

Dmrecord 250436

Document Type Dissertation

Rights Lynch, Kirsten Mary

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA