University of Southern California Dissertations and Theses

High resolution elastography in ophthalmology

(USC Thesis Other)

High resolution elastography in ophthalmology

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

High Resolution Elastography in Ophthalmology

by

Runze Li

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOMEDICAL ENGINEERING)

August 2022
ii

Dedication

For my beloved
Parents
Dr. Changhe Li and Yali Hou

Runze Li

iii

Acknowledgments
This thesis and the related projects can not be made without the help and support from
the entire team. This journey toward the Ph.D. degree has brought me so many warm and
wonderful people, I would like to express my gratitude to the whole team.
Foremost, I would like to thank my advisor, Dr. Qifa Zhou, from the deep of my heart
for his uncompromising support throughout my whole doctoral research at USC. His
insightful advice has always been inspiring me, his attitude to the hard work has been
pushing me to overcome the difficulties, and his encouragement has been delivering the
enthusiasm to never give up. I learned not only the skills, knowledge, and professions from
him, but most importantly, his life philosophy about how to be a good person truly
influenced me.
I would like to thank the committee members of my dissertation, Dr. Mark S. Humayun,
Dr. Keyue Shen, Dr. Cristina Zavaleta, and Dr. Benjamin Xu for the valuable suggestions,
and the effort they spent to improve the quality of the dissertation.
I would like to thank Dr. Mark S. Humayun and Dr. Zhongping Chen for their consistent
support in our collaboration projects over the years. The guidance and mentoring from
them that I received were tremendous, as I were like one of their students. Thanks a lot for
sharing your expertise with me.
Over the years in Zhou’s lab, I was lucky to be able to work with those smart and
warmhearted senior members. I wish to acknowledge Dr. Ruimin Chen, Dr. Teng Ma, Dr.
Yang Li, Dr. Harry Chiu, Dr. Mingyue Yu, Dr, Zeyu Chen. The research work could be
very tough without their help. I would like to thank our visiting scholars who have been
sharing their expertise from other fields with me: Dr. Jun Zhang, Dr. Zhiqiang Zhang, Dr.
Tianfu Zhang, Dr. Di Li, Dr. Hanmin Peng, Dr. Laiming Jiang, Dr. Yi Quan, and Dr. Jiapu
Li.
Our current lab members have been supporting me and encouraging me, especially
during the pandemic time, such a special period. I would like to acknowledge to: Dr.
Xuejun Qian, Dr. Robert Wodnicki, Dr. Haochen Kang, Gengxi Lu, Yizhe Sun, Junhang
Zhang, Adnan Rayes, Yushun Zeng, Chen Gong, and Dongliang Yan. Thanks for being
not only my colleagues but my friends. The memories that we all shared and the friendship
iv

that we have will be an invaluable fortune in my future. In particular, I would like to express
my gratitude to Dr. Xuejun Qian for the generous help and unflappable support over the
years.
I would like to acknowledge the help and support from our IBT members who made the
animal study possible. My special thanks to Dr. Ying Liu for his generosity of willingness
to help at any time and for sharing his expertise with me.
Last but not least, I would like to acknowledge the support and accompany from my
friends, thanks to my parents for their unconditional love, without whom this work would
not be possible.

Table of Contents

Dedication .......................................................................................................................... ii
Acknowledgments ............................................................................................................ iii
Abstract .............................................................................................................................. v
Chapter 1 Introduction..................................................................................................... 1
1.1 Biomechanics in Ocular Tissue ................................................................................. 1
1.2 Ultrasonic and optical coherence elastography ......................................................... 2
1.2.1 Excitation method ............................................................................................... 2
1.2.2 Imaging Methods ................................................................................................ 3
1.3 Motivations and Objectives ....................................................................................... 6
1.4 Outline ....................................................................................................................... 7
Chapter 2 Ultrasonic elastography to assess biomechanical properties of the optic
nerve head and peripapillary sclera of the eye ............................................................. 10
2.1 Introduction ............................................................................................................. 10
2.2 Methods ................................................................................................................... 12
2.2.1 Experimental setup ........................................................................................... 12
2.2.2 Post-processing and data analysis..................................................................... 14
2.2.3 Phantoms and biologic tissue preparation ........................................................ 15
2.3 Results ..................................................................................................................... 16
2.3.1 Phantom results................................................................................................. 16
2.3.2 ONH and PPS response to intraocular pressure ............................................... 17
2.3.3 Statistical analysis............................................................................................. 18
2.4 Discussion and Conclusion ..................................................................................... 19
Chapter 3 Quantitative confocal optical coherence elastography for evaluating
biomechanics of optic nerve head using Lamb wave model ....................................... 23
3.1 Introduction ............................................................................................................. 23
3.2 Methods ................................................................................................................... 25
3.2.1 Preparation of Specimens ................................................................................. 25
3.2.2 System Setup .................................................................................................... 25
3.2.3 Lamb Wave Model ........................................................................................... 28
3.2.4 Post-processing ................................................................................................. 28

3.3 Results ..................................................................................................................... 29
3.3.1 Calibration of the Imaging System ................................................................... 29
3.3.2 ONH Response to Elevated IOP ....................................................................... 30
3.4 Discussion and Conclusion ..................................................................................... 32
Chapter 4 In vivo evaluation of posterior eye elasticity using shaker-based optical
coherence elastography .................................................................................................. 35
4.1 Introduction ............................................................................................................. 35
4.2 Methods ................................................................................................................... 37
4.2.1 Experimental setup ........................................................................................... 37
4.2.2 Post-processing and data analysis..................................................................... 38
4.2.3 Phantoms and rabbit preparation ...................................................................... 39
4.3 Results ..................................................................................................................... 39
4.3.1 Phantom results................................................................................................. 39
4.3.2 In vivo rabbit posterior eye ............................................................................... 41
4.4 Discussion and Conclusion ..................................................................................... 43
Chapter 5 High Resolution Optical Coherence Elastography of Retina under
Prosthetic Electrode ........................................................................................................ 46
5.1 Introduction ............................................................................................................. 46
5.2 Methods ................................................................................................................... 48
5.2.1 Experimental setup ........................................................................................... 48
5.2.2 Post-processing and data analysis..................................................................... 49
5.2.3 Implant surgery ................................................................................................. 49
5.3 Results ..................................................................................................................... 50
5.3.1 Prosthetic electrode characterization ................................................................ 50
5.3.2 Surgery characterization ................................................................................... 51
5.3.2 Biomechanical response of retina to the prosthetic electrode .......................... 52
5.4 Discussion and Conclusion ..................................................................................... 55
Chapter 6 Simultaneous Assessment of the Whole Eye Biomechanics Using
Ultrasonic Elastography ................................................................................................. 63
6.1 Introduction ............................................................................................................. 63
6.2 Methods ................................................................................................................... 65
6.2.1 Experimental Set-up ......................................................................................... 65
6.2.2 Data Processing ................................................................................................ 66

6.2.3 Phantom Preparation and System Synchronization .......................................... 67
6.2.4 Animal Protocol ................................................................................................ 68
6.3 Results ..................................................................................................................... 69
6.3.1 System Validation with Phantom Study ........................................................... 69
6.3.2 Results for Rabbit Eye ...................................................................................... 71
6.4 Discussion and Conclusion ..................................................................................... 75
Chapter 7 Research the Anisotropy of the Equatorial Sclera with Ultrasonic
Elastography .................................................................................................................... 79
7.1 Introduction ............................................................................................................. 79
7.2 Methods ................................................................................................................... 80
7.3 Results ..................................................................................................................... 81
7.4 Discussion and Conclusion ..................................................................................... 93
Chapter 8 Summary and Future Work ........................................................................ 95
8.1 Summary ................................................................................................................. 95
8.2 Future Work ............................................................................................................ 96
8.2.1 2D Array-based Elastography System .............................................................. 96
8.2.2 Human Study .................................................................................................... 97
Bibliography .................................................................................................................. 100

List of Figures
Figure 1-1. Overview of eye’s anatomy with key ocular components labelled. (Ethier et
al. 2004, Annu Rev Biomed Eng.)
............................................................................................................................................. 2
Figure 1-2. Overview of excitation methods in elastography. (Kennedy et al. 2017, Nat
Photonics.)
............................................................................................................................................. 3
Figure 1-3. (a) Frequency response spectrogram of agar and metal ball. (b) 3D OCE
image. (c) The sample image
............................................................................................................................................. 4
Figure 1-4. Schematics of the (i) unloaded and (ii) preloaded setup. (b,c) Corresponding
OCT B-scans (d) Stress-strain curve of the sensor material. (e) B-scan view of the
displacement due to actuation. (f) Corresponding strain in the sample. (Kennedy et al.
2015, Sci. Rep.)
............................................................................................................................................. 5
Figure 1-5. Representative post-processing method. (a) In time domain. (b) In frequency
domain. (Shih et al. 2018, Ieee T Med. Imaging.)
............................................................................................................................................. 6
Figure 2-1. Schematic diagram of the shaker induced ultrasonic elastography system and
the synchronized timing sequence
........................................................................................................................................... 13
Figure 2-2. The spatiotemporal displacement map of (a,b) the homogenous phantoms
and (c) the side-by-side phantom. SWSs is the shear wave speed of soft phantom while
SWS. (h) indicates the stiffer phantom. The inflection point in (c) is the boundary
between the soft and stiff part of the side-by-side phantom
........................................................................................................................................... 16
Figure 2-3. The B-mode image and the reconstructed Young’s modulus SWE map of the
homogenous phantoms with, (a,b) soft phantom, (c,d) stiff phantom, and (e,f) the side-by-
side phantom
........................................................................................................................................... 16
Figure 2-4. The B-mode image and the reconstructed Young’s modulus map of the
posterior pole of the porcine eye including ONH and PPS under five different IOPs – 6
mmHg, 12 mmHg, 18 mmHg, 24 mmHg and 30 mmHg. The second column shows the
posterior pole biomechanics in an increased IOP while the third column shows that in a
decreased IOP
........................................................................................................................................... 17
Figure 2-5. The relationship between the biomechanics (Young’s modulus) of the ONH
and PPS under different IOPs
........................................................................................................................................... 19
Figure 3-1. Optoacoustic elastography setup with unscalded porcine eye. SLD,
superluminescent diode; OC, optical coupler; CO, collimator; OA, optical attenuator; M,
mirror; GM, galvanometer mirrors; L1/L2: lens; UT, ultrasonic transducer; RFA,

radiofrequency amplifier; FG, function generator; and G, grating. (b) Schematic of the
IOP control and measurement system
........................................................................................................................................... 26
Figure 3-2. Schematic representation for synchronization of the complete imaging
system for synchronized confocal OCE. M-mode displacement map data were acquired in
each location (e.g., P0, P1, P2, etc.)
........................................................................................................................................... 27
Figure 3-3. (a) The OCT intensity image of the posterior segment of the eye. (b)–(d) A
time series of ODT raw data shows the elastic wave propagating across the ONH. Axial
displacement is related to color, as shown in the color bar at the right of the figure
........................................................................................................................................... 30
Figure 3-4. Data postprocessing flow during each biomechanics assessment with curves
generated by averaging axial displacement through the entire thickness with lateral
positions: (a) spatial–temporal map, (b) k-space by 2-D FFT, (c) frequency normalized by
corresponding maximum wavenumber, (c) phase velocity, and (d) curve fit in the
frequency of 0 to 1200 Hz in 125-Hz increments
........................................................................................................................................... 31
Figure 3-5. Experimental phase velocity (diamond symbols) and corresponding Lamb
wave model curve fit with different IOP levels: (a) 10 mmHg, (b) 20 mmHg, and (c) 30
mmHg. Error bars represent the standard deviation between experimental data and fitted
curve at each sampled frequency
........................................................................................................................................... 32
Figure 3-6. The averaged Young’s modulus of the ONH with elevated IOP. Error bars
represent the deviation for each measurement. The curve fitting results show good
agreement with a polynomial dependence to IOP
........................................................................................................................................... 32
Figure 4-1. Shaker-based OCE system schematic diagram with in vivo rabbit eye setup.
SLD: superluminescent diode: OC: optical coupler: CO: collimator: OA: optical
attenuator: M: mirror: GM: galvanometer mirrors: L1/L2: lens: RFA: radiofrequency
amplifier: FG: function generator: G: grating
........................................................................................................................................... 37
Figure 4-2. Displacement curves of the homogeneous phantom. (a) With 200 µ s, 600 µ s,
and 1000 µ s shaker pulse duration at the initial galvo position (the most left position of
the OCT image). (b) At the depth of 50 µm, 250 µm, and 450 µm under the condition of
600 µ s shaker pulse duration. (A color version of this figure is available in the online
journal.) ............................................................................................................................. 40
Figure 4-3. (a) OCT image of the phantom, (b) spatiotemporal displacement map of the
homogeneous phantom, (c) shear wave propagation at the timing of 0, 0.088, 0.176 ms,
respectively
........................................................................................................................................... 41
Figure 4-4. (a) OCT image of the posterior rabbit eye in vivo, (b–d) shear wave
propagation maps at the timing of 0, 0.088, 0.176 ms, respectively. Layer 1: Nerve fiber,
ganglion cell, and inner plexiform; Layer 2: inner nuclear, outer plexiform, and outer

nuclear; Layer 3: RPE; Layer 4: choroid; Layer 5: sclera
........................................................................................................................................... 42
Figure 4-5. Spatiotemporal displacement maps of different layers of the posterior eye. (a)
Layer 1, (b) Layer 2, (c) Layer 3, and (d) Layer 5
........................................................................................................................................... 42
Figure 5-1. Characterizations of retinal prosthetic electrode. (A) Structural image of the
prosthesis; (B,C,D) optical images of the flexible prosthesis when it stands freely, bends
convexly and concavely with more than 90°
........................................................................................................................................... 51
Figure 5-2. a-c Cross-sectional OCT images of retina with prosthetic electrode. d-e
RetCam images of retina after phacoemulsification procedure and implantation procedure
respectively. ONH: optic nerve head, RV: retina vessels
........................................................................................................................................... 52
Figure 5-3. Spatial-temporal maps in each layer of retina with and without prosthetic
electrode. The slope of the white line represents the corresponding elastic wave speed.
Color bar represents the axial displacement
........................................................................................................................................... 54
Figure 5-4. Statistical analysis of each layer in the retina with and without prosthetic
electrode
........................................................................................................................................... 54
Figure 6-1. Schematic diagram of the shaker based ultrasonic elastography system for
the in vivo study
........................................................................................................................................... 65
Figure 6-2. Schematic diagram of the IOP control and monitoring system
........................................................................................................................................... 66
Figure 6-3. (a) Schematic diagram of the phantom study; (b) The synchronization
sequence of ultrasonic elastography system
........................................................................................................................................... 68
Figure 6-4. Phantom study results to validate the accuracy of the ultrasonic elastography
system. (a) B-mode imaging and the corresponding spatial-temporal map. The dashed
line represents the wavefront of the elastic wave; (b) The propagation of the elastic wave
in the phantom with a time interval of 1 ms. The dashed line represents the wavefront of
the elastic wave, the wave propagates from left to right. Axial displacement unit: µ m
........................................................................................................................................... 70
Figure 6-5. Spatial-temporal maps of the cornea, iris, lens, PPS and ONH at the IOP of
10 mmHg. Dashed lines represent the wavefront of the elastic wave
........................................................................................................................................... 71
Figure 6-6. Statistical analysis of the Young’s modulus distributions of the ocular tissues
with different IOP levels
........................................................................................................................................... 72
Figure 6-7. The Young’s modulus distributions of the ocular tissues with different IOP
levels and the corresponding fittings. The dots represent the experimental results, the

dashed lines represent the fittings according to the experimental results
........................................................................................................................................... 73
Figure 6-8. B-mode imaging and the elastic wave velocity mapping of the same eye
under different IOP levels. The first column represents the B-mode imaging and the
second column represents the corresponding mapping
........................................................................................................................................... 74
Figure 7-1. (a) The diagram of the high frequency ultrasonic elastography configuration.
(b) the locations of the superior, inferior, nasal and temporal quadrants
........................................................................................................................................... 81
Figure 7-2. B-Mode Imaging and Spatial-Temporal Map in Different IOP at Superior
along equatorial direction
........................................................................................................................................... 82
Figure 7-3. B-Mode Imaging and Spatial-Temporal Map in Different IOP at Superior
along AP direction
........................................................................................................................................... 83
Figure 7-4. Spatial-Temporal Map at two directions in Different IOP at Temporal
quadrant............................................................................................................................. 84
Figure 7-5. Spatial-Temporal Map at two directions in Different IOP at Nasal quadrant.
........................................................................................................................................... 85
Figure 7-6. Spatial-Temporal Map at two directions in Different IOP at Inferior quadrant.
........................................................................................................................................... 86
Figure 7-7. The biomechanics mapping at the superior quadrant with two directions
under different IOP levels. For better visualization and enhance the image contrast, the
elastic wave speed is used instead of the Young’s modulus. The color bar represents the
elastic wave speed, unit is m/s
........................................................................................................................................... 87
Figure 7-8. The biomechanics mapping at the temporal quadrant with two directions
under different IOP levels
. ......................................................................................................................................... 88
Figure 7-9. The biomechanics mapping at the nasal quadrant with two directions under
different IOP levels.
.......................................................................................................................................... 89
Figure 7-10. The biomechanics mapping at the inferior quadrant with two directions
under different IOP levels.
........................................................................................................................................... 90
Figure 7-11. Statistical analysis of the Young’s modulus for all the measurements. The
dashed line represents the second polynomial fitting, error bar represents the standard
derivation.
........................................................................................................................................... 91
Figure 7-12. Statistical analysis of the anisotropy in different quadrants under each IOP
level. The dashed line represents the second polynomial fitting.
........................................................................................................................................... 92

Figure 8-1. Verasonics system and a 2D ultrasonic array transducer. (Source data from
Verasonics website)
........................................................................................................................................... 97
Figure 8-2. Homemade 12 MHz high frequency 2D array for elastography application.
........................................................................................................................................... 97
Figure 8-3. The design of screening stage, 10 degrees of freedom.
........................................................................................................................................... 98
Figure 8-4. (a)B-mode imaging of the anterior eye, (b) a typical screening scenario, (c)
spatial-temporal map of my iris.
........................................................................................................................................... 99

v

Abstract
Elastography is a widely used imaging modality to assess the biomechanical properties
of tissue in a non-invasive manner, evidence shows that with the progression of certain
diseases, the morphology changes are indiscernible, but the biomechanical changes are
prominent, hence, elastography provides additional diagnosis information besides the
conventional structural imaging.
Recently, elastography in ophthalmology is growing rapidly, for it can reconstruct
the biomechanics non-invasively while maintain the original structure of the imaging
object. Palpation, air puff, acoustic micro-tapping, acoustic radiation force (ARF) and
shaker have been used to excite the ocular tissue and generate tissue motion, ultrasound
and optical coherence tomography (OCT) have been used to track this tissue motion and
reconstruct the ocular tissue biomechanics. However, palpation is limited by user-
dependence, while air puff and acoustic micro-tapping suffer from the energy
attenuation from tissue, those methods are not suitable for the posterior segment of the
eye. Furthermore, most conventional ultrasonic elastography studies carry out a standard
frequency range, which can not meet the need for high spatial resolution in ocular tissue.
The work presents in this dissertation proposal develops single element ultrasonic
transducer based elastography, high frequency ultrasonic array based elastography, and
optical coherence elastography (OCE) and investigates their applications in ophthalmology.
One of the advantages that these systems have in common is the high resolution. This
advantage enables to reconstruct the biomechanical properties of the imaging target
accurately. In this study, the imaging or detecting method includes single element
ultrasonic transducer, array transducer and OCT. Single element transducer based
ultrasonic elastography is more capable of in vitro study. Comparing with the array
transducer, the performance of the single element transducer is typically better owing to
the manufactural difficulty of the array, for example, single element transducer gains better
sensitivity and the bandwidth. However, single element transducer based ultrasonic
elastography requires mechanical scanning during the measurement. The slow imaging
speed restricts the application of this system into clinic, its high performance imaging
transducer makes it more suitable for in vitro study. OCE has the highest resolution among
vi

these imaging systems, it also gains the advantage of the imaging the posterior eye due to
the transparency of the cornea. It is capable of imaging one specific ocular tissue, but it can
barely build the biomechanical connections with more ocular tissues due to the limited
field of view.
Due to the regulation from U.S. Food and Drug Administration (FDA) on the ultrasound
exposure in the eye, a shaker has been used instead of ARF to generate the tissue motion.
After calibrated with imaging phantoms of our imaging systems, the biomechanics
mappings of optic nerve head (ONH) and peripapillary sclera (PPS) have been provided
by a single element transducer elastography system, then an advanced Lamb wave model
has been applied with OCE system to reconstruct the biomechanical properties of ONH
with more accuracy. Furthermore, the OCE system has been applied in vivo to reconstruct
the biomechanics of the retina in living rabbits, and the biomechanical effects of the retinal
prosthetic electrode on the retina have been further investigated. Finally, the quantitative
biomechanics mappings of the in vivo rabbit eyes have been reconstructed under different
intraocular pressure (IOP) levels with the array based elastography system. The anisotropy
of the equatorial sclera has been revealed experimentally with the same system. These
studies demonstrate the feasibility of high resolution elastography in the applications of
ophthalmology.

1

1. Chapter 1 Introduction

1.1 Biomechanics in Ocular Tissue
As a complex organ that conveys and decode light into neural stimuli, the eye consists
of several interactive components, such as cornea, sclera and ONH as shown in Figure 1-
1. Dysfunction of the eye is usually associated with alteration of the biomechanical
property. For example, keratoconus is a prevalent disease that leads to significant visual
impairment due to the development of a cone-shaped ectatic cornea, and one of its clinical
signs is an unusually compliant cornea compared with normal cornea [1]; Glaucoma is a
leading cause of irreversible blindness, elevated intraocular pressure (IOP) is assumed to
be the primary risk factor, ONH tends to be stiffer with the progression of glaucoma [2].
Age-related macular degeneration (AMD) is one of the leading causes of severe,
irreversible vision loss in people aged over 60 [3]. It has been shown that the main cause
of visual loss in AMD is the development of choroidal neovascularization (CNV). the
mechanical properties of distinct cellular layers in the retina are altered with the onset of
AMD [4]. Along with the conventional structural imaging methods, there is in need to
provide additional diagnostic imaging modalities in ophthalmology.
The tissue stiffness can be quantified by Young’s modulus (𝐸 ). The Young’s modulus
is correlative with the ratio of the stress (𝜎 ) and the corresponding strain (𝜀 ). Stress
represents the force per unit area that counteracts the applied force, and the strain relates
the deformed configuration of a material to its initial reference configuration.
The Young’s modulus is defined using Eq-1
𝐸 =
𝜎 𝜀 =
𝐹 𝐴 ⁄
∆𝐿 𝐿 ⁄
Eq-1
where 𝐸 is the Young’s modulus, 𝐹 is the applied force, 𝐴 is the surface area, ∆𝐿 is the
tissue displacement, 𝐿 is the original length.
When assuming the elastic wave propagates in a bulk, homogeneous, isotropic and
purely elastic medium, the Young’s modulus can be reconstructed using Eq-2
𝐸 = 3𝜌 𝐶 𝑠 2
Eq-2
2

where 𝜌 is the medium density, 𝐶 𝑠 is the group velocity of the induced elastic wave.

Figure 1-1. Overview of eye’s anatomy with key ocular components labelled. (Ethier et al. 2004, Annu Rev
Biomed Eng.)

1.2 Ultrasonic and optical coherence elastography
Elastography is an emerging imaging modality which can interpret the biomechanical
properties of soft tissue from strain, elasticity and viscosity aspects in a non-invasive
manner. In the tissue and organ levels, there are magnetic resonance elastography (MRE),
conventional UE, OCE and newly developed high frequency UE available to characterize
the biomechanical properties, among which, MRE and UE provide the resolution above
the milliliter range, while high frequency UE and OCE provide micrometer or sub-
micrometer resolution. Owing to the subtle structure of the eye, high frequency UE and
OCE are preferred to characterize the biomechanical properties of the ocular tissue.

1.2.1 Excitation method
Acoustic radiation force (ARF) is a physical phenomenon resulting from the acoustic
wave propagation interacting with a dissipative medium. It is caused by the transfer of
3

energy from the acoustic wave to the subject, arising from absorption and scattering. In
soft tissue, the acoustic impedance matches with the water interface, hence, the
contribution of scattering is negligible to ARF. Thus, the magnitude of this force can be
defined as Eq-3:
𝐹 =
2𝛼𝐼
𝑐 Eq-3
where 𝐹 is acoustic radiation force, 𝛼 is the absorption coefficient of the medium, 𝑐 is the
speed of sound in the medium, 𝐼 is the temporal average intensity at a given point in space
[13, 14]. For a focused beam, ARF is distributed in the focal area, this area is called region
of excitation (ROE). The elastic wave propagates orthogonally with the direction of ARF,
starting from the edge of the ROE.
Using ARF as the excitation method has the advantages of dynamic focusing and broad
bandwidth of induced elastic wave, however, it is not appropriate for clinical trials in
ophthalmology due to the FDA regulation of ultrasound exposure. Shaker is an optimal
replacement in this case to generate the tissue motion and initiate the elastic wave
propagation. It touches the eye gently, exerts the force directly on the ocular tissue, and
has been applied in a human study [5].

1.2.2 Imaging Methods
Both UE and OCE mainly contain three categories as shown in Figure 1-2: 1) harmonic
elastography, 2) quasi-static compression elastography, 3) transient or impulse
elastography. The principles underlying these three categories are different:

Figure 1-2. Overview of excitation methods in elastography. (Kennedy et al. 2017, Nat Photonics.)

1) Harmonic elastography:
4

A continuous sinusoidal acoustic wave is applied into the sample either localized to a
point or across the whole sample, and the resonance frequency is determined
experimentally with scanning load frequency when the max tissue displacement occurs.
Under Hook’s law, the Young’s modulus is positively related to the square of the resonance
frequency. Agar phantom is much more compliant compared with the metal ball, Figure 1-
3 shows the principle of concept and proves it in the harmonic elastography: agar has the
greater axial displacement and lower resonance frequency. However, this method is being
questioned that the measured resonance frequency is more likely to be affected by the
structure and boundary geometry, not the mechanical properties.

Figure 1-3. (a) Frequency response spectrogram of agar and metal ball. (b) 3D OCE image. (c) The sample
image.

2) Compression elastography:
The concept of compression elastography is to utilize the definition of Young’s modulus.
The external force is applied to the tissue sample by either piezo actuator or manual
palpation, the strain corresponding to the same force is acquired by either OCT or
ultrasonic transducer, strain imaging can be provided in a 3D volumetric manner. Moreover,
a calibration layer that can convert the strain into stress, provides the Young’s modulus
mapping throughout the sample tissue. This method has the advantages such as fast
5

imaging speed, however, a flat surface of sample tissue is required or preload is needed,
which limits its application significantly.

Figure 1-4. Schematics of the (i) unloaded and (ii) preloaded setup. (b,c) Corresponding OCT B-scans (d)
Stress-strain curve of the sensor material. (e) B-scan view of the displacement due to actuation. (f)
Corresponding strain in the sample. (Kennedy et al. 2015, Sci. Rep.)

3) Transient or impulse elastography
The tissue sample is excited by a transient pulse acoustic wave, the induced elastic
wave propagates orthogonally with the force direction. With the assumptions for the
tissue sample of bulk, homogenous, pure elastic and isotropic, the shear modulus is
6

positively related to the square of the induced wave speed. With the known Poisson ratio,
Young’s modulus is approximately three times the shear modulus. This method gains
many of the advantages, such as dynamic excitation of the region of interest and flexible
post-processing. Fig 1-4. (a) and (b) shows the reconstruction of the mechanical
properties in time domain and frequency domain respectively. In this proposal, we mainly
use this method to excite sample tissue and initiate the elastic wave propagation.

Figure 1-5. Representative post-processing method. (a) In time domain. (b) In frequency domain. (Shih et
al. 2018, Ieee T Med. Imaging.)

1.3 Motivations and Objectives
Due to the subtle structure of the eye, there is a high demand for high-resolution
imaging modalities in ophthalmology. Most ultrasonic elastography study is based on
commercial ultrasonic transducer or transducer array, the spatial resolution ranges from
milliliter to several milliliters, and restricts its imaging applications in both research and
clinical ophthalmology. Owing to the nature of the high resolution of OCT and the
transparency of the ocular tissue, OCT is a promising imaging modality to serve as a
diagnostic tool in ophthalmology. As the detection imaging modality, high frequency
ultrasound and OCT are capable of its high resolution in the ophthalmic study. For the
excitation, harmonic elastography could make bias estimation of the biomechanical
properties due to the complex structure and interface medium in the eye, while applying
a certain preload to make the eye in a flat shape is not acceptable, these facts indicate
that the combination of the impulse force as the excitation method, high frequency
ultrasound and OCT as the detection method is capable to characterize the
7

biomechanical properties of the eye, and have the potential to be translated into clinic
as a routine diagnostic tool.

1.4 Outline
The thesis proposal is outlined as follows:
Chapter 1 introduces the concept of soft tissue biomechanics and its clinical requirement
for elasticity imaging. Then, a more controllable and accurate excitation method called
acoustic radiation force (ARF) was introduced. The ARF-based ultrasonic elastography
including ARFI and SWEI is proposed to quantify tissue biomechanics in both qualitative
and quantitative ways. Finally, the motivation and potential clinical needs for developing
high frequency ultrasound based elastography technique are addressed.
Chapter 2 utilizes high resolution ultrasonic elastography to quantitatively evaluate the
biomechanical properties of both optic nerve head (ONH) and peripapillary sclera (PPS)
in porcine eyes in response to both increasing and decreasing IOP. Homogeneous phantoms
and side-by-side imaging phantoms (to mimic the anatomical structure of ONH and PPS)
were first used to test the accuracy of our imaging system. Then both point-to-point
Young’s modulus maps and statistical analysis were provided at various IOP levels.
Chapter 3 investigates the reconstruction of ONH’s biomechanical properties with our
confocal OCE system and an advanced Lamb-wave model. The ONH exhibits a think plate
structure, its whole thickness is comparable with the wavelength of the shear wave, group
velocity is not suitable to be used. Given the facts that: 1) the elastic wave propagation
medium (ONH) is in a thin plate shape, 2) the boundary interfaces of this medium are
incompressible (fluid), 3) ARF is generated orthogonal with the elastic wave propagation
directions, an antisymmetric Lamb wave is determined to be the induced elastic wave. The
Doppler OCT data of ONH in various IOP levels is first acquired with our confocal ARF-
OCE system, and the biomechanical properties are reconstructed with the Lamb wave
model.
Chapter 4 utilized shaker based OCE to investigate the biomechanical properties of
posterior eye in vivo. Different with the previous chapter, a shaker is used as the vibration
source instead of ARF, due to the strict regulation of the U.S. FDA on the ultrasound
8

exposure in ophthalmology. A homogenous phantom study is conducted to assess the
accuracy of our system and optimize our system as well. The biomechanical properties of
posterior eye in vivo are evaluated.
Chapter 5 presents high resolution OCE of the retina under the prosthetic electrode.
Retinal prosthetic electrode holds the potential to restore some of the visual functions. It
has been investigated with the safety and bio-compatibility aspects, however, how it
changes the retinal biomechanics remains unknown. Shaker based OCE technology is used
to measure the retinal biomechanics before and after prosthetic implantation, further
supplementing data for a potential clinical trial.
Chapter 6 demonstrates the reconstruction of the biomechanics of the major ocular
tissues in the living rabbit which covers the whole eye range under different IOP levels.
The eye is a highly interactive organ, some diseases such as glaucoma could affect the
biomechanics of the eye systematically. However, how these major ocular tissues interact
with the fluctuation of the IOP remains unknown. An ultrasonic array based elastography
has been developed, the accuracy has been verified, the biomechanics of the major ocular
tissues have been reconstructed, both the statistical analysis and the mappings have been
provided in this study for the first time. It has the potential to be translated into the clinic
as a routine diagnostic method.
Chapter 7 explored the anisotropy properties of the equatorial sclera. The biomechanics
of the equatorial is important as it serves as the protective layer for the retina to resist the
internal IOP fluctuation and prevent the external trauma, it also relates to the progression
of myopia. However, the research of on biomechanics of the equatorial sclera is limited.
Anisotropy in this study is defined as the difference in biomechanics caused by different
directions. This study decodes the anisotropy property of the equatorial sclera
quantitatively and experimentally. It could not only facilitate the research about how the
progression of glaucoma changes the anisotropy, why the myopic eye tends to be elongated
in the axial direction instead of the azimuthal direction, it could also give a hint to making
the adhesives for ocular trauma with more biocompatibility.
Chapter 8 summarizes the current work on high frequency ultrasonic elastography and
OCE, and the applications in ophthalmology. The future work will focus on the translation
of the array based ultrasonic elastography to the clinic, investigate the changes in
9

biomechanics between normal people and glaucomatous patients. 2D ultrasonic array for
the 3D volumetric elastography will also be conducted, it has the advantage of being faster,
more accurate, it could be a powerful tool when assessing the anisotropy properties by
controlling the vibrating and imaging receiving directions.

10

2. Chapter 2 Ultrasonic elastography to assess biomechanical
properties of the optic nerve head and peripapillary sclera
of the eye
2.1 Introduction
Glaucoma is a leading cause of irreversible blindness, and it is estimated that by 2020
the number of people suffering glaucoma will reach 80 million worldwide[6]. Although
elevated intraocular pressure (IOP) is the primary risk factor for the development of
glaucoma, the mechanisms by which elevated IOP eventually leads to damage and loss
of neural function are still unclear [2]. It is also unclear how sensitivity to IOP varies
and interacts with other risk factors for glaucoma. The optic nerve head (ONH) is the
primary site of damage in glaucoma, and the mechanical properties of the adjacent
peripapillary sclera (PPS) are known to strongly influence the stresses and strains of the
ONH where the retinal ganglion cell axons exit the eye[7, 8]. Specifically, elevated IOP
may cause the scleral canal opening to widen circumferentially and the ONH to bow
posteriorly, resulting in axon dysfunction, death, or activation of detrimental cellular
response[9, 10]. It is therefore important to characterize the mechanical properties of the
posterior eye under various IOP conditions[11].
Due to the location in the posterior segment of the eye, it is difficult to investigate the
biomechanical properties of ONH and PPS in a non-invasive manner. Tensile testing
[12, 13] and atomic force microscopy (AFM) nanoindentation [14] are two commonly
used methods to measure the biomechanical properties of the ONH and PPS. Both
methods require targeted tissue to be dissected from the eye globe into stripes, after
which cryosections are further prepared for AFM testing. To maintain the complex
physiological loading conditions of the eye in its three dimensional configuration, in
vitro inflation testing of the posterior sclera tissue has been developed as an
improvement to the uniaxial strip test methodology [15]. However, with this technique,
the deformation force must still be manipulated by time dependent IOP changes
invasively, which impedes potential application. Moreover, these techniques destroyed
the original boundary conditions of the ocular tissues, which were likely to affect the
accuracy of the measurements.
11

Elastography is an emerging imaging modality to quantify biomechanics in soft tissue,
mostly in a non-invasive manner. For ophthalmology, the more often reported
elastography techniques such as magnetic resonance elastography (MRE) [16] and
conventional ultrasound elastography (UE) [17] are not suitable mainly because of their
low spatial resolution. Recently developed optical coherence elastography (OCE)
techniques, utilizing optical coherence tomography (OCT) [18] to detect the propagation
of induced elastic waves, have wide application in characterizing biomechanics in
cornea [19], lens [20] and retina [21]. However, owing to the limited penetration depth,
OCE systems lack the ability to image the deeper ONH region underneath the lamina
cribrosa and non-transparent sclera tissue. It is notable that high frequency ultrasound
has become an indispensable technique for ophthalmic imaging owing to its natural
advantage of balanced spatial resolution and penetration depth [22, 23].
To investigate the biomechanical properties of posterior eye, especially for IOP-
related deformation, some research studies have developed ultrasound imaging-based
methods. Specifically, Alam et al. [24] implemented the sonoelastic Doppler ultrasound
method to investigate the relations of mechanical resonance frequency (which is directly
related to tissue elasticity) of the eye with elevated IOP using various models. Pavlatos
et al. [25] developed a high frequency ultrasound speckle tracking method to investigate
regional deformation of the ONH and PPS during IOP elevation on ex vivo porcine eyes.
Later, Ma et al. [26] implemented the same technology using human donor globes.
However, in their setup, the displacement vector must be calculated between images
acquired at different IOP levels (i.e., the deformation force was generated by the
changing IOP), which may not be feasible in practice. To generate a more controllable
and stable deformation, an external pushing force generated by an air-pulse, acoustic
radiation force (ARF) or mechanical shaker is preferred. Among these pushing strategies,
the mechanical shaker is preferred given the requirement for generating sufficient
deformation in posterior locations and also due to potential safety issues [5]. In addition
to the pushing method, previously reported results only present either regional
displacements or strain, which are all qualitative parameters of the tissue biomechanics.
Without calibration of the applied force, the absolute Young’s modulus or elasticity
cannot be conclusively determined.
12

In this study, we developed and implemented a mechanical shaker based ultrasonic
elastography technique that has the capability to provide a quantitative estimation of
tissue biomechanical properties via shear wave elasticity (SWE) imaging technology
with high spatial resolution. The goal of this study is to quantify and map the mechanical
properties of both ONH and PPS in response to either increasing or decreasing IOP for
the purpose of better understanding the mechanism of development of Glaucoma.

2.2 Methods
2.2.1 Experimental setup
A schematic diagram of the experimental setup and synchronized timing sequence is
shown in Fig 2-1. Due to strict U.S Food and Drug Administration (FDA) 510 k
standards for ophthalmic exposure, a mechanical shaker (mini-shaker type 4810; Bruel
& Kjaer, Duluth, Georgia, USA) was used here as the pushing source in place of the
low frequency high power ultrasound transducer used in our previous studies [22, 27,
28]. To precisely track tissue motion and elastic wave propagation caused by the
mechanical shaker, a 40 MHz needle transducer was designed and fabricated in this
study.

13

Figure 2-1. Schematic diagram of the shaker induced ultrasonic elastography system and the synchronized
timing sequence.
To acquire SWE imaging, the high frequency needle transducer was mounted on a 3-
axis translation motorized linear stage (SGSP33-200, OptoSigma Corporation, Santa
Ana, CA, USA) for mechanical scanning. Owing to the small size of the needle
transducer aperture which is less than 0.4 mm in diameter, the pushing force generated
by the mechanical shaker is applied almost orthogonally to the imaging subject, meaning
only shear waves are generated. In order to track shear wave propagation, the shaker
was fixed at the target position while the needle transducer was controlled by the
motorized linear stage which was moved based on the designed distance between the
positions of the mechanical shaker and needle transducer. To sufficiently cover the ONH
and PPS region, the step size and scanning distance were set to 42 µ m and 8.4 mm,
respectively. In addition, to avoid any issues with vibration of the needle during
movement, the time delay between successive positions was set to 100 ms, including the
data acquisition time and extra wait time.
During the experiment, an arbitrary function generator (AFG 3252C, Tektronix,
Beaverton, OR, USA) generating a single pulse signal with 500 µ s pulse duration was
connected to a power amplifier (Type 2718, Bruel & Kjaer, Duluth, Georgia, USA)
which was itself coupled to transmit an amplified signal to the mechanical shaker to
induce tissue deformation. It is important to note that an impulsive pushing force is
benefit to generate broadband shear wave while reducing the deformation of the tissues
[29]. Thus, based on our previous study [30], 500 µ s pulse duration was selected here as
a balance between the high temporal resolution (transient pushing) and stability of the
shaker (i.e., too short pulse duration will cause incorrect pushing pattern of the shaker).
In addition, the pulse amplitude was set to 600 mV in order to induce sufficient pushing
force which can ensure that the displacement of the shear wave was detectable within
the imaging region of the ONH and PPS. The needle transducer was set in conventional
pulse-echo mode and was driven by a pulser/receiver (JSR500, Ultrasonics, Pittsford,
NY, USA) with a pulse repetition frequency (PRF) of 20 kHz. After 20–80 MHz analog
band-pass filtering (Mini-Circuits, Brooklyn, NY, USA) to remove signal contamination,
14

the radiofrequency (RF) ultrasonic data was captured using a 12-bit digitizer (ATS9360,
Alazartech, Montreal, QC, Canada) at a sampling rate of 1.8 GHz and stored for off-line
analysis. To reduce system jitter, the same clock was used to synchronize the digitizer,
pulser/receiver, and arbitrary function generator. To record the initial tissue position at
each scanning position, the mechanical shaker was excited 50 µ s after the needle
transducer started to acquire data. At each tracking location, a total of 100 A-lines were
acquired for each M-Mode image. All parameter settings were kept constant for all
measurements.

2.2.2 Post-processing and data analysis
Data analysis was performed using MATLAB 2018a software (The MathWorks,
Natick, MA, USA). At each M-mode, the first tracking A-line was served as the
reference, and then dynamic tissue displacements were calculated using 1-D normalized
cross-correlation [31] with a symmetric search region and 1.5λ window size (λ is the
wavelength of the tracking transducer) between the reference A-line and the rest of A-
lines in the M-mode. By repeating such procedures along lateral positions, we achieved
a 3D matrix (expressed as lateral positions, depth positions and time) with displacement
values as its index.
As demonstrated by previous studies [22, 32], using a small propagation distance to
reconstruct the local shear wave speed (SWS) is preferred to sustain a high resolution
SWS map. However, both ONH and PPS are associated with high elasticity, especially
at high IOP. Therefore, in this study, a relatively long lateral propagation distance
interval (756 µ m) was selected to quantify each pixel of the SWS image mapping with
a good balance between precision and resolution.
More specifically, we first converted the 3D matrix into the 2D spatiotemporal map
(propagation distance versus time shifts curve) at each depth position where the wave
propagation distance was measured by accumulation of lateral positions and the time
shift was defined as the time to reach the peak displacement at each dynamic
displacement. Then, the original spatiotemporal map was 2D interpolated to a finer grid
size using a spline function. Next, the SWS at each pixel of image was estimated by
15

applying a linear regression to a subset of 756 µ m interval (lateral propagation distance)
in the interpolated spatiotemporal map. By repeating the linear regression procedures to
these subsets with a 42 µ m (the moving step size of the needle transducer) distance shift
each time, we achieved SWS image mapping. The final reconstructed SWS speed
images were generated after applying a 3 × 3 median filter so as to increase the signal-
to-noise-ratio (SNR). To quantify the Young’s modulus imaging mapping, we used the
equation (1-2) to convert SWS to Young’s modulus in each pixel.

2.2.3 Phantoms and biologic tissue preparation
Gelatin (Gelatin G8-500, Fisher Scientific, USA) based tissue-mimicking phantoms
with equal concentration of silicon carbide powder (S5631, Sigma-Aldrich, St. Louis,
MO, USA) and sound scatters were fabricated. The stiffness of each phantom was
controlled by gelatin concentration. Three phantom types were made, including two
homogeneous phantoms with different stiffness and a side-by-side phantom with two
different stiffness levels in a single structure. The stiffness of the two homogeneous
phantoms was tested using the gold standard – uniaxial mechanical testing (Model 5942,
Instron Corp., MA, USA), and the results showed 8.1 ± 0.8 kPa for the soft phantom and
25.4 ± 1.9 kPa for the stiff phantom (the mean and standard deviation were determined
by measuring each phantom five times), respectively.
Eight fresh, unscalded porcine eyes were obtained from a local service-oriented
company (Sierra Medical Science, Inc., Whittier, CA, USA) within 24 h of death in this
study. The extraocular tissues were trimmed away, and the optic nerve was cut to be
flush with the outer surface of the PPS. Eyes were mounted on a customized holder with
the ONH and PPS exposed on the top. The lens and vitreous were removed and two ports
were inserted into the eye chamber through the corneal limbus. One port was connected
to the balanced saline solution bag set at various heights to manipulate IOP. The other
port was connected to a pressure sensor (Model SPR-524, Millar Inc, TX, USA) to read
the true IOP inside the chamber. In this study, five IOPs at both increasing and
decreasing levels were investigated, including 6 mmHg, 12 mmHg, 18 mmHg, 24 mmHg
and 30 mmHg.
16

2.3 Results
2.3.1 Phantom results
The tissue mimicking phantoms were first used to verify the stability and accuracy of
the shaker-induced ultrasonic elastography system. Fig 2-2 (a-b) shows the spatiotemporal
displacement maps of the two homogeneous phantoms at a certain depth location
(randomly selected depth at the region of interest). As can be observed in Fig 2-2, owing
to homogeneity of the phantom, constant generated SWS resulted in a linear relationship
between the shear wave propagation distance and the propagation time. Using linear
regression of the data in the spatiotemporal displacement maps, the SWS of the soft
phantom and stiff phantom were calculated to be 1.6 ± 0.1 m/s, 2.8 ± 0.3 m/s, respectively.
The corresponding Young’s modulus values calculated using Eq-2 are consistent with the
gold standard value reported above, and demonstrate that our imaging system has the
ability to accurately determine the absolute Young’s modulus. In addition to showing
spatiotemporal information calculated at a certain depth, the uniform color mapping
depicted in Fig 2-3 (b,d) also indicates consistency of the obtained Young’s modulus values
along the depth direction as expected for the two homogenous phantoms.

Figure 2-2. The spatiotemporal displacement map of (a,b) the homogenous phantoms and (c) the side-by-
side phantom. SWSs is the shear wave speed of soft phantom while SWS. (h) indicates the stiffer phantom.
The inflection point in (c) is the boundary between the soft and stiff part of the side-by-side phantom.

Figure 2-3. The B-mode image and the reconstructed Young’s modulus SWE map of the homogenous
phantoms with, (a,b) soft phantom, (c,d) stiff phantom, and (e,f) the side-by-side phantom.
17

2.3.2 ONH and PPS response to intraocular pressure
Fig 2-4 shows the 2D cross-sectional B-mode images and corresponding SWE images
of the posterior segment of a porcine eye. A total of five different IOP levels were
investigated, including one low IOP (6 mmHg), two normal physiological IOP (12 mmHg
and 18 mmHg) and two high IOP (24 mmHg and 30 mmHg).

Figure 2-4. The B-mode image and the reconstructed Young’s modulus map of the posterior pole of the
porcine eye including ONH and PPS under five different IOPs – 6 mmHg, 12 mmHg, 18 mmHg, 24 mmHg
and 30 mmHg. The second column shows the posterior pole biomechanics in an increased IOP while the
third column shows that in a decreased IOP.

The differences in the appearance of the B-mode images obtained at each IOP level are
barely discernable (i.e., subtle changes in thickness or curvature). In other words, it is
difficult to provide quantitative measurements by relying solely on the B-mode images. In
examining Fig 2-4 where the color-coded Young’s modulus in the SWE images represents
stiffer regions in red and softer areas in blue, the intensity of SWE imaging increases
18

rapidly with IOPs, which indicated that SWE imaging is far more sensitive to changes in
IOP than gray-scale B-mode. In addition, both ONH and PPS have a tendency to become
stiffer at a higher IOP. This change in elasticity was noted to be more prominent in the PPS
than it was in the ONH.

2.3.3 Statistical analysis
To statistically determine the relationship between the Young’s modulus of ONH / PPS
with increasing / decreasing IOP, eight porcine eyeballs were used. For each eyeball, we
calculated one value for ONH and one for PPS. More specifically, the average Young’s
modulus of the central ONH contributed to one value. The left PPS and right PPS were
averaged to represent the overall Young’s modulus for the PPS. The reconstructed Young’s
modulus values were all expressed as mean ± standard deviation (eight eyes).
With increasing IOP, the estimated Young’s modulus values for the PPS were 176.8 ±
14.3 kPa at 6 mmHg, 201.3 ± 13.2 kPa at 12 mmHg, 230.6 ± 16.3 kPa at 18 mmHg, 451.5
± 100.5 kPa at 24 mmHg, and 573.5 ± 64.4 kPa at 30 mmHg, respectively. With respect to
the ONH, the corresponding reconstructed Young’s modulus values were 92.4 ± 13.9 kPa
at 6 mmHg, 112.7 ± 18.6 kPa at 12 mmHg, 134.9 ± 18 kPa at 18 mmHg, 158.3 ± 22.8 kPa
at 24 mmHg, and 224.7 ± 71.1 kPa at 30 mmHg, respectively. It is generally observed that
PPS is stiffer than ONH and the stiffness of PPS increases more rapidly than that of ONH.
For decreasing IOP, the reconstructed Young’s modulus of PPS and ONH were 662 ±
78 kPa / 344.7 ± 58 kPa at 30 mmHg, 480 ± 12.7 kPa / 253.2 ± 2.9 kPa at 24 mmHg, 392.4
± 27.5 kPa / 244.6 ± 23.3 kPa at 18 mmHg, 297.5 ± 17 kPa / 198.7 ± 27.3 kPa at 12 mmHg,
237.3 ± 5.6 kPa / 157.9 ± 13.6 kPa at 6 mmHg.
The differences between the various IOP levels were evaluated by one-way ANOVA.
The statistical analysis showed that changes in Young’s modulus were statistically
significant at different IOP values at either increased or decreased IOP. In this study, P less
than 0.05 was considered to be a significant difference. In Fig 2-5, the averaged Young’s
modulus of ONH and PPS at different IOP levels are plotted. Due to the large tissue
deformation variation and non-linearity, we fit the data using a 2nd order polynomial
function as opposed to a linear fitting function.
19

Figure 2-5. The relationship between the biomechanics (Young’s modulus) of the ONH and PPS under
different IOPs.

2.4 Discussion and Conclusion
We have presented the mapping of stiffness for both ONH and PPS for increasing and
decreasing IOP using a shaker-induced ultrasonic elastography system. We used a high
frequency and broad bandwidth needle transducer, and therefore our system has the ability
to detect micrometer-level axial displacement and provide imaging mapping with high
spatial resolution. The high spatial resolution of our system enables the acquisition of
accurate spatial–temporal maps for subsequent post-processing. The deep penetration
depth of the ultrasound signal covers the whole posterior segment of the eye, which enables
the elastography system to provide a comprehensive stiffness mapping of the ONH and
PPS.
To precisely induce localized shear wave propagation, most previous ultrasound
elastography systems or optical coherence elastography systems utilized either an air pulse
or acoustic radiation force generated by an ultrasonic transducer as an external pushing
source. However, both of these are not suitable in this study, because the air pulse approach
has a low bandwidth of induced mechanical waves and slow relaxation times [29], while
there is strict FDA regulation of ultrasound in ophthalmology (i.e., to induce a detectable
deformation on ONH and PPS, a large power is essentially required to be applied on the
ultrasound transducer, resulting in a significantly higher mechanical index and acoustic
intensity). Therefore, a mechanical shaker which has previously been investigated on
20

human subjects clinically [5] was preferred in this study to vibrate the tissue to initiate
elastic wave propagation. The effect of different pushing duration was discussed in our
previous work [30]. In this study, 500 µ s pulse duration was used as well to ensure a long
propagation distance of the induced elastic wave under the transient pushing.
Two homogeneous tissue mimicking phantoms and a homogeneous side-by-side
phantom were first used to validate the feasibility of our imaging system. By applying
linear regression to a mapping of the spatiotemporal positions, Young’s modulus values at
each spatial domain were reconstructed. The reconstructed Young’s modulus values via
shear wave speed were comparable and matched the gold standard mechanical test. All
these results demonstrate that the shaker-induced ultrasonic elastography system can
provide reliable measurement of the elasticity distribution of the imaged subject, which is
in a similar geometry in ONH and PPS (i.e., side-by-side structure).
Investigation of biomechanical properties such as Young’s modulus of ONH and PPS
with elevated IOP has been conducted by other groups through tensile testing [12] and by
using an AFM [14]. More specifically, a range from 10 to 36 MPa and ~17 kPa stiffness
level have been reported, respectively. It should be noted that both ONH and PPS need to
be cut into stripes for the tensile test, and micro-meter level specimens are required in AFM
testing. In other words, the difference of geometry in these studies results in the large range
of the measured Young’s modulus. To better describe the stiffness mapping of the ONH
and PPS, non-invasive imaging of the target region of interest (ROI) in an intact eye
structure is preferred. Pavlatos et al. [25] and Ma et al. [26] implemented an ultrasound
speckle tracking method to measure the deformation of the posterior segment of porcine
and human eyes, respectively. However, in their studies, only strain imaging was provided,
and they therefore lacked quantitative measurement of the Young’s modulus. By contrast,
our previous OCE work on the backside of the porcine eye [33] has shown that the Young’s
modulus of ONH ranges from 180 kPa to 450 kPa with an increasing IOP from 10 mmHg
to 30 mmHg. Despite the fact that PPS measurements were lacking with our OCE system,
the measured Young’s modulus values for the ONH are comparable with the results in this
study.
Jan et al. quantified the collagen crimp or micrometer-scale waviness of lamina
cribrosa (LC) and PPS with varying IOP [34-36]. Their studies experimentally revealed
21

that collagen fiber waviness decreased with elevated IOP, which in turn caused the fibers
to straighten and stretch to become recruited, leading to stiffening of the tissue. Our
experimental results which showed that tissue elasticity went up with the increasing of IOP,
were consistent with those previous studies. In addition, our findings indicated that both
ONH and PPS become stiffer when IOP drops from a high pressure. It is possible because
that the tissue collagen crimp does not return to its original form when IOP returns to its
initial level following elevation. As a consequence, the overall posterior segment of the eye
at decreased IOP becomes stiffer than that of increased IOP.
Moreover, collagen fibers uncramp when stretched, which implies that the posterior
ocular tissues are initially compliant at low stretch level but increase rapidly in stiffness at
a higher deformation level [37]. The non-linear relationship between ocular tissue stretch
and IOP elevation has previously been observed experimentally in the posterior sclera and
ONH [38, 39]. In this study, we successfully demonstrated the trend of Young’s modulus
of ONH and PPS with varying IOPs, and confirmed its non-linear properties. However, the
2nd order polynomial function fitting used in this study might not be rigorous because of
only 8 eyes involved. More samples in the future study would be helpful to develop a more
accurate relationship between Young’s modulus of ONH/PPS and IOPs.
There are a few limitations in this study. First, despite the good agreement between
the reconstructed Young’s modulus and gold standard in the phantom study, the group
shear wave speed (using Eq-2 to reconstruct the Young’s modulus) based reconstruction
method may not be accurate, and may lead to some bias in the ocular tissue studies because
of boundary conditions and the ratio of ONH/PPS thickness to the shear wavelength. In
addition, viscosity is another parameter that is crucial for biomechanical characterization
of tissues, including ONH and PPS. A detailed analysis of mechanical mode propagation
in a bounded medium (i.e., ONH and PPS) is preferred, however, it is beyond the scope of
this manuscript. Future studies will develop advanced models to quantify both tissue
elasticity and viscosity by evaluating dispersion of shear wave propagation. Second, our
study shows that both ONH and PPS tend to be stiffer with increased IOP rather than
decreased IOP. This is because the collagen fibers remain straight and therefore lead to
local stiffening. Since only 8 porcine eyes were investigated, more samples should be tested
22

in the future in order to obtain a comprehensive understanding of stiffness variations of the
ONH and PPS with glaucoma.
In summary, we have shown proof of principle of using a shaker-induced ultrasonic
elastography system to characterize biomechanical properties of the ONH and PPS
quantitatively via the direct indicator: Young’s modulus. The performance of our imaging
system was first calibrated in homogeneous, side-by-side gelatin tissue mimicking
phantoms, then tested on porcine eyeballs ex vivo. With the IOP elevation, the
reconstructed Young’s modulus of ONH raised from 92.4 ± 13.9 kPa at 6 mmHg to 224.7
± 71.1 kPa at 30 mmHg non-linearly, while PPS showed a similar tendency from 176.8 ±
14.3 kPa at 6 mmHg to 573.5 ± 64.4 kPa at 30 mmHg. It is generally observed that at the
same IOP level, the posterior segment of the eye at decreased IOP is stiffer than that of
increased IOP. Overall, simultaneously investigating the link between the biomechanical
properties of the ONH / PPS with both increased and decreased IOP may provide new
insights to better understand the progression of the glaucoma. The proposed ultrasonic
elastography method might be a powerful tool to assess both morphological and
biomechanical properties of ocular tissue.
23

3. Chapter 3 Quantitative confocal optical coherence
elastography for evaluating biomechanics of optic nerve
head using Lamb wave model
3.1 Introduction
As an optic neuropathy, glaucoma is one of the leading causes of blindness
worldwide[40, 41]. Glaucomatous damage to the visual system likely includes important
pathophysiologies within the photoreceptors, retinal ganglion cell (RGC), lateral
geniculate body, and visual cortex. Evidence has shown that damage to the RGC axons
within the lamina cribrosa (LC) of the optic nerve head (ONH) is a central
pathophysiology[42].
The ONH is a region of special biomechanical interest because it is a discontinuity in
the cornea–scleral shell, and this type of discontinuity typically gives rise to stress or strain
concentrations in mechanical systems [43, 44]. The mechanical properties of the ONH are
dependent on the interaction of its individual components. At low stress, it is likely that the
elastin component allows for an initially large volume change formed by the distending
ONH and, as pressure increases and the extended collagen fibers limit further deformation,
the ONH becomes progressively stiffer [45]. Elevated intraocular pressure (IOP) is the
most important risk factor for glaucoma progression [46], which distorts the tissues of the
ONH, and LC within, triggering events, such as compromised axoplasmic flow, vascular
perfusion, and astrocyte activation that eventually lead to optic nerve axon degeneration
and RGC death [46, 47]. An improved understanding of the ONH biomechanical
environment and of the dependence of this environment on the biomechanical properties
of the ONH tissues is necessary to understand better how biomechanical effects may play
a role in glaucomatous optic neuropathy [8].
There are several experimental techniques currently being used to investigate changes
in the mechanical compliance of the ONH and posterior scleral surface [15, 25]. Most of
these identify the pressure-induced deformation at the level of the LC and use this
information to determine the biomechanical properties of the tissue [25]. However, it is
known that the ONH and anterior laminar surfaces are not only displaced posteriorly but
also anteriorly (with reference to Bruch’s membrane opening) in a significant portion of
24

glaucoma patients [48]. Characterizing elasticity of the ONH over physiological pressures
may provide a better understanding of how changes in IOP lead to changes in the
mechanical environment of the ONH [7, 49, 50].
Optical coherence elastography (OCE) is an emerging technique, which can detect tissue
biomechanics noninvasively with both high temporal and spatial resolution compared with
conventional ultrasonic elastography [17]. The axial displacement range is typically in the
micrometer to nanometer level. Recent work on advanced ultrasonic elastography
demonstrated excellent spatial resolution with micrometer axial displacement detection [22,
27]. OCE shows great potential to assess the biomechanics of the posterior segment of the
eye. In ophthalmology, however, this technique is mainly focused on the anterior segment
of the eye, such as the cornea and the lens [29, 51-53]. Some OCE systems utilize oblique
incidence air pulses [51, 54] or an obliquely oriented transducer [55] to induce tissue
vibration, which may generate complex waves instead of pure waves and can lead to errors
in the estimation of biomechanical parameters [56]. We recently reported a confocal
acoustic radiation force (ARF) OCE method that can map out the elasticity of retinal layers
with high resolution [57]. This confocal setup avoids generating complex waves, moreover,
it allows for easy access to the retinal layers. However, due to the complexity of the retina,
only group velocity was used to estimate the biomechanics of normal retina and age-related
macular degeneration-induced retina.
One of the challenges in elastography is to reconstruct the biomechanics accurately [58].
Group velocity has long been used to reconstruct the biomechanics, however, this method
assumes that the elastic wave propagates in a homogeneous, isotropic, bulk, and purely
elastic medium [59]. In most tissues, elasticity and viscosity exist at the same time.
Moreover, some tissues such as the ONH exhibit a thin plate morphology, which leads to
a failure of biomechanics estimation using group velocity.
In our previous work, we demonstrated the feasibility of a Lamb wave model to quantify
the viscoelasticity of imaging phantoms, porcine corneas as well as rabbit arteries [28]. In
this study, we investigate the ability of OCE for measurement of the ONH elasticity in pig
eyes, utilizing a pressure inflation setup in which IOP is controlled precisely. We further
utilize the Lamb wave model to fit the phase dispersion curve during our data
postprocessing procedure. To our knowledge, this is the first time that a Lamb wave model
25

has been used in evaluating the biomechanics of the ONH. Our aim was to quantify the
mechanical behavior of the ONH in response to elevated IOP.

3.2 Methods
3.2.1 Preparation of Specimens
Twenty porcine eyes were obtained from a U.S. Department of Agriculture-approved
slaughterhouse. The mixed-breed pigs were all healthy and between 5 and 7 months old
before they were killed. The eyes were immersed in saline and kept cold during shipping,
and all eyes were tested within 48 h postmortem. The extraocular tissues were removed,
and the optic nerve was cut to be flush with the outer surface of the peripapillary sclera by
using a surgical blade. At the ONH, the axons converge into bundles to pass through the
LC, which spans the scleral canal and is continuous with the adjacent peripapillary sclera.
Viewed from the back of eyeball, the ONH was surrounded by peripapillary sclera at the
same level after the optic nerve was removed. Then, the entire eyeball was physically
stabilized in a custom holder with the ONH facing upward. Two ports were created in the
pars plana using 23 G trocars (BL5523WVX, Bausch & Lomb, Inc., Rochester, New York).
One connected to a pressure column containing phosphate-buffered saline (PBS) to control
IOP, and the other connected with an IOP sensor (Model SPR-524, Millar, Inc., Texas).
Real-time readout of the IOP was performed using a custom-build LabView program,
which provided feedback on the instantaneous IOP. The height of the infusion line was
adjusted accordingly. The sample was then put in a PBS bath, which served as a medium
for ultrasound propagation and a preservative for ocular tissue. The IOP was increased by
manually raising the height of the PBS column from 10 to 30 mmHg. We used a pressure
increment of 5 mmHg for each step, and the IOP was held constant at each level for 1 min
before the data were acquired

3.2.2 System Setup
To assess the biomechanical properties of the ONH, a customized OCE system
combining an ultrasonic transducer and a spectral-domain optical coherence tomography
26

(SD-OCT) system was developed, as shown schematically in Fig 3-1. (a) In the SD-OCT
system, a light beam was emitted from a superluminescent diode (SLD, M-D-890-HP1,
Superlum Diodes Ltd, Carrigtwohill, Ireland) with a central wavelength of 890 nm and a
bandwidth of 144 nm to detect tissue structural information as well as response to
ultrasonic ARF. The light beam was filtered through an optical isolator (IO-F-SLD150-
895, Thorlabs, Inc., Newton, New Jersey) to prohibit backprojection of light into the SLD
and then split by an 80/20 optical coupler for efficient and safe data collection. In the
reference arm, the light beam was collimated, attenuated, and reflected. In the sample arm,
light was collimated, scanned through a galvanometer (GVS002, Thorlabs Inc.), and then
focused via a 54-mm scan lens (LSM04-BB, Thorlabs Inc.) through the hollow part of the
ring transducer to the tissue. The backscattering signal from the sample arm was interfered
with the reflected signal from the reference arm and transmitted to the detecting arm. This
interference signal was split by wavelength via a diffraction grating and finally detected by
a line scan CCD camera (spL4096-140 km, Basler AG, An der Strusbek 60–62, Germany)
at a 50-KHz A-line rate. The signal was further processed to OCT intensity images and
phase-resolved displacement maps. The axial resolution of this SD-OCT is around 3μm in
air.

Figure 3-1. Optoacoustic elastography setup with unscalded porcine eye. SLD, superluminescent diode; OC,
optical coupler; CO, collimator; OA, optical attenuator; M, mirror; GM, galvanometer mirrors; L1/L2: lens;
UT, ultrasonic transducer; RFA, radiofrequency amplifier; FG, function generator; and G, grating. (b)
Schematic of the IOP control and measurement system.
A custom-built 5 MHz modified PZT material ring transducer was used to generate
tissue motion. The aperture size is 23 mm with a f-number of 1. The peripapillary sclera
27

was selected as the ultrasonic excitation area, where uniform distribution of acoustic output
initiated the propagation of elastic waves in the tissue. The detected displacement of the
ONH was attributed to the response of this generated ARF. The excitation pulse was
limited to 2 ms to allow the tissue to respond to the ARF naturally and recover to its original
position before the next ultrasonic excitation. Confocal alignment of the ultrasound
transducer and the SD-OCT system was achieved by using hydrophone (HGL-0085,
ONDA Co, Sunnyvale, California) guidance before the experiment.
To acquire the OCT intensity images and phase-resolved displacement maps, a global
clock was used to synchronize all subunits of the overall imaging system, including the
ultrasound transducer, the CCD imager, and the SD-OCT, as shown in Fig 3-2. ARF was
generated with the galvanometer set to the P0 location and then, the elastic wave was free
to propagate along with the imaged object. M-mode raw data were acquired at each of the
galvanometer locations (e.g., P0, P1, P2, etc.) for 400 A-lines resulting in a total location
interval of 8.8 ms (0.02 ms delay in every A-line for data storage). The step size for
consecutive lateral positions was 4μm with 1000 lateral points in total. Then, the raw data
were saved to disk for offline processing.

Figure 3-2. Schematic representation for synchronization of the complete imaging system for synchronized
confocal OCE. M-mode displacement map data were acquired in each location (e.g., P0, P1, P2, etc.).

28

3.2.3 Lamb Wave Model
The Young’s modulus can be estimated simply in free space using the Eq-2 [22].
The shear modulus (
μ
) can be obtained using the density (
ρ
) and constant group velocity
(
𝑐 𝑔
). The Young’s modulus (E) is approximately three times the shear modulus. One of the
assumptions above also includes a pure elastic medium for guided wave propagation.
However, the ONH exhibits a thin-layer structure and is small compared with the shear
wavelength. Giving that lamb waves typically travel in thin plates with a fluid boundary
interface and ARF is applied orthogonally, a lamb wave is identified as the guided wave
propagated in the ONH. The speed of Lamb waves is dispersive and they can only
propagate in the low frequency range due to the viscosity in soft tissue. Since ARF was
generated vertically from top to bottom through the thickness of the fluid column, with
particles oscillating in the same direction, anti-symmetric lamb waves are mainly induced.
Moreover, the 0th order mode can travel at any frequency and is most commonly detected
in the low frequency range as compared to non-0th order mode lamb waves. For a thin
plate submerged in an incompressible fluid, the dispersion equation based on the 0th order
anti-symmetric lamb wave mode is as follows [59]:

4𝑘 𝐿 3
𝛽 𝐿 cosh 𝑘 𝐿 ℎ sinh 𝛽 𝐿 ℎ − 𝑘 𝑠 2
− 2𝑘 𝐿 2

2
sinh 𝑘 𝐿 ℎ cosh 𝛽 𝐿 ℎ = 𝑘 𝑠 4
cosh 𝑘 𝐿 ℎ cosh 𝛽 𝐿 ℎ
Eq-4
Where
𝛽 𝐿
= 𝑘 𝐿 2
−𝑘 𝑠 2

,
𝑘 𝐿 = 𝜔 /𝑐 𝐿
is the Lamb wave number,
𝜔
is the angular
frequency,
𝑐 𝐿
is the frequency dependent phase velocity,
𝑘 𝑠 = ω 𝜌 / 𝑈
is the shear
wavenumber,
U = μ + 𝑖 𝜔 𝜂
is the shear modulus,
μ
and
𝜂
are shear elasticity and shear
viscosity respectively,
ℎ
is the half thickness of the ONH.

3.2.4 Post-processing
The OCT intensity and Doppler phase raw data were first obtained, the spatial-
temporal map along with lateral positions and time was then acquired by averaging the
axial displacements through the entire thickness of the ONH. The ultrasonic excitation
region in the field of view were all within the full width at half maximum axial intensity of
29

acoustic field. The group velocity was obtained by applying linear regression to the wave
crest at different lateral positons. The spatial-temporal map was then transformed to
frequency / wavenumber domain (k-space map) by using the 2D Fast Fourier Transform
(FFT). The k-space map below 5 Hz was masked to avoid low frequency noise. The phase
velocity was acquired by dividing the frequency by the corresponding maximum
wavenumber as follows:
𝑐 𝐿 = 2𝜋 𝑓 /𝑘 𝐿
Eq-5
The phase velocity was sampled up to 1200 Hz in 125 Hz increments. Then the shear
modulus and viscosity-related fitted phase velocity were selected to minimize the error in
every sampled phase velocity
𝑅 2

using the following cost function [59]:
𝑅 2
= 𝑐 𝑓 𝜇 ,𝜂 −𝑐 𝐿 𝑓 1
,𝑓 2
,… ,𝑓 𝑛
2

Eq-6
Where
𝑐 𝑓
is the curve fitted phase velocity, f1...fn are sampling frequencies with 125 Hz
interval. The error rate was also calculated to evaluate the precision of the fitting procedure
by using experiment data and fitted data at every sampled frequency [28].

𝐸 𝑅 = 𝐴 𝑣 𝑒 𝑟 𝑎 𝑔 𝑒

𝑐 𝑓 𝑓 −𝑐 𝐿 𝑓
2
𝑐 𝐿 𝑓

Eq-7

3.3 Results
3.3.1 Calibration of the Imaging System
The acoustic field of the 5 MHz ultrasonic transducer was measured using a needle
hydrophone aligned to the focal point of the transducer. The measured -6 dB lateral beam-
width was 384 µ m, the -6 dB axial focal distance was 3.84 mm which can cover the OCT
imaging depth in the ONH (~ 0.4 mm) and enable an even distribution of acoustic intensity
during data acquisition. In a previous study, the reconstructed Young’s modulus of an agar
imaging phantom using group velocity was found to be comparable to mechanical testing
[22]. We also previously demonstrated the feasibility of reconstructing the Young’s
modulus of rabbit cornea and the posterior segment of the eye in vivo [20, 57]. With respect
30

to the Lamb wave model, we previously showed the accuracy of this model in calculating
the Young’s modulus in a thin and viscoelastic gelatin phantom, as well as for corneal and
arterial tissues [28].

3.3.2 ONH Response to Elevated IOP
The biomechanical response of the ONH to elevated IOP was investigated with our
Lamb wave model. After each data acquisition, OCT intensity and phase-resolved Doppler
OCT data were acquired. Figure 3-3 (a) shows the OCT intensity image of the posterior
segment of the eye. By reslicing the ODT raw data, the elastic wave is shown as
propagating along this area. Figure 3-4 (a)–(d) show results on each postprocessing
procedure. By averaging over the entire thickness direction of axial displacements along
with lateral positions and time, the spatial–temporal map was generated. The wave crest in
different lateral positions exhibits a linear trend, and the group velocity can be estimated
using a linear regression algorithm. The 2-D FFT of the spatial–temporal map was then
used to generate the k-space map as a function of frequency and wavenumber. The phase
velocity curve was then obtained using Eq-5, with the frequency at each point divided by
the maximum intensity at that frequency. Due to the nature of viscosity in the ONH, the
phase velocity is dispersive, as shown in Figure 3-4 (d). The frequency range was up to
1200 Hz. Finally, the phase velocity curve was sampled and fitted to the Lamb wave model
using Eq-4 with minimized error Eq-6.

Figure 3-3. (a) The OCT intensity image of the posterior segment of the eye. (b)–(d) A time series of ODT
raw data shows the elastic wave propagating across the ONH. Axial displacement is related to color, as shown
in the color bar at the right of the figure.

31

Figure 3-4. Data postprocessing flow during each biomechanics assessment with curves generated by
averaging axial displacement through the entire thickness with lateral positions: (a) spatial–temporal map,
(b) k-space by 2-D FFT, (c) frequency normalized by corresponding maximum wavenumber, (c) phase
velocity, and (d) curve fit in the frequency of 0 to 1200 Hz in 125-Hz increments.

Figure 3-5. (a-c) shows the sampled phase velocity and curve fitting results as
described above for a range of IOP levels, where IOP = 10 mm-Hg, 20 mm-Hg and 30
mm-Hg respectively. The experimental data matches the curve-fit result with good
agreement, and the average ERs are from 5.4% to 11.09% when the IOP is below 30 mm-
Hg. The Young’s modulus can be obtained by this curve-fitting procedure and is shown in
Figure 3-6: 181.3 ± 21.0 kPa at IOP of 10 mm-Hg, 210.0 ± 22.5 kPa at IOP of 15 mm-Hg,
231 ± 31.6 kPa at IOP of 20 mm-Hg, 324.8 ± 41.3 kPa at IOP of 25 mm-Hg, 450.5 ± 52.5
kPa at IOP of 30 mm-Hg respectively. The Young’s modulus of ONH can be curve-fit to
the corresponding IOP using a piecewise polynomial. The goodness-of-fit was 99.7%
which suggests that it should be possible to evaluate Young’s modulus with higher IOP in
future work.

32

Figure 3-5. Experimental phase velocity (diamond symbols) and corresponding Lamb wave model curve fit
with different IOP levels: (a) 10 mmHg, (b) 20 mmHg, and (c) 30 mmHg. Error bars represent the standard
deviation between experimental data and fitted curve at each sampled frequency.

Figure 3-6. The averaged Young’s modulus of the ONH with elevated IOP. Error bars represent the deviation
for each measurement. The curve fitting results show good agreement with a polynomial dependence to IOP.

3.4 Discussion and Conclusion
In this paper, we have presented reconstruction of the Young’s modulus of the ONH
by combining our OCE system with a Lamb wave model for the first time. The high spatial
and temporal resolution of our OCE system enables the acquisition of a highly sensitive
spatial–temporal map, which enables high accuracy for subsequent postprocessing. The
confocal configuration of the transducer with OCT allows the ARF to be applied to the
imaging object orthogonally, which avoids generation of complex elastic waves, such as
Love waves, which increases the accuracy of biomechanical estimation. The confocal
configuration also allows the OCT beam to penetrate through the center hole in the
33

ultrasonic transducer without incurring significant attenuation, which should enable us to
assess the biomechanics of the ONH in vivo for future studies. We further introduced a
Lamb wave model in this study. By fitting the experimental phase velocity to the Lamb
wave model with low ER, the biomechanics of the ONH can be reconstructed with elevated
IOP.
Previous work by other groups studied the Young’s modulus of the LC and
parapapillary sclera as a measure of ONH stiffness [12-14]. A common approach for the
determination of Young’s modulus is tensile testing of tissue strips [12, 13]. However,
different techniques of measuring stiffness give different stiffness values, making a direct
comparison of these values difficult. For instance, a stress–strain measurement showed
Young’s modulus of the LC was in the range of E = 11.8 to 15.6 MPa and E = 28.5 to 36.0
MPa of the parapapillary sclera [12], whereas a Young’s modulus of 17.2 ( ± 2.7) kPa was
measured for the normal human LC using atomic force microscopy nanoindentation [14].
To the best of our knowledge, this study is the first to apply OCE to analyze Young’s
modulus as a measure of stiffness in the ONH. As expected, Young’s modulus measured
reflects the more real values representative of whole eyes.
Although the quantification of the ONH biomechanical properties with elevated IOP
has been demonstrated, there are still some challenges that need to be addressed for future
applications. First, a faster imaging speed is required, especially for stiffer imaging objects.
For this study, the imaging speed is sufficient to track the elastic wave propagation in the
ONH. But for the peripheral sclera, which is stiffer than the ONH, our imaging system may
not be fast enough to track wave propagation. A highspeed camera could help solve this
issue. Second, the measured mechanical index of the current system is around 1.5, which
exceeds U.S. Food and Drug Administration (FDA) regulations (0.23 in ophthalmology).
Since, in this study, we acquired only a few hundred nanometers of axial displacement [60,
61], the ultrasonic output can be reduced to meet the FDA regulations and maintain the
ability to track elastic wave propagation. Third, ER is increased with elevated IOP. Since
ultrasonic output remains unchanged in this study, the ONH became stiffer with elevated
IOP, thus, the axial displacement decreased correspondingly. The signal-to-noise ratio may
be decreased with a lower detected axial displacement, which leads to a high ER.40
However, the estimated elasticity is still in a reasonable range. Finally, we showed that it
34

is feasible to reconstruct Young’s modulus more accurately using ONH as the subject,
however, the viscosity of the ONH with elevated IOP is not provided here. Tissue is
predominantly elastic, with viscosity being typically much smaller as compared with
elasticity. According to Eq. (2), when
μ ≫ 𝜔 𝜂
, the contribution of viscosity to overall
viscoelasticity decreases as elasticity increases, curve fitting is unaffected by changes in
viscosity. Moreover, the viscosity of the ONH is not important in this study, and to the best
of our knowledge, there is currently no disease related to abnormal viscosity or alteration
of viscosity.
In summary, we have demonstrated a method to quantify the biomechanics of the ONH
with elevated IOP. The new method uses a confocal OCE system and realizes both high
temporal and spatial resolution. We explored the feasibility of applying the Lamb wave
model to reconstruct the biomechanics of thin-layered and viscoelastic ONH. By fitting the
phase dispersion curve with minimized , the biomechanics of the ONH can be
accurately determined. The study suggests that it should be feasible to evaluate
biomechanics in vivo using the confocal configuration, which could lead to assessment of
the biomechanics of the ONH in the clinic, allowing us to study the progression of
glaucoma over time. Due to this advance, we believe that our confocal OCE system and
methodology holds great potential for the detection of biomechanical properties in the
posterior segment of the eye in clinical diagnosis.

35

4. Chapter 4 In vivo evaluation of posterior eye elasticity using
shaker-based optical coherence elastography
4.1 Introduction
Retinal diseases, such as age-related macular degeneration (AMD) are the leading cause
of severe, irreversible vision loss in people aged over 60 [3]. It has been shown that the
main cause for visual loss in AMD is the development of choroidal neovascularization
(CNV). Although there are available treatments for CNV, they are not capable of reversing
the injury nor able to improve vision in the majority of cases [62]. As a result, early
detection is the key to preservation functional vision.
To understand the pathogenesis of the AMD, several ocular imaging technologies have
been developed for the purpose of AMD detection and monitoring. The first transformative
imaging was the invention of fluorescein angiography to visualize the vessel
distribution/blood leakage if neovascularization is suspected [63]. Later, another imaging
modality namely fundus photography has been developed on epidemiological studies of
AMD to establish major risk factors [64, 65]. More recently, a new imaging modality
referring to optical coherence tomography (OCT) [66] has been introduced to provide
better visualization of the various retinal layers non-invasively. Despite these techniques
are capable of providing crucial information for the diagnosis of AMD, they often are
insufficient for early diagnosis, before structural changes occur. It recently has been shown
that the mechanical properties of distinct cellular layers in the retina are altered with the
onset of AMD [67]. Thus, investigating the biomechanical properties of the posterior eye,
especially retina, is essential for understanding its physiological function and their response
to stress from injuries, medical device implantation, and potential surgeries.
Elastography, an imaging modality capable of mapping the biomechanical properties of
soft tissues, provides additional contrast mechanism and clinically relevant information for
disease diagnosis. A typical elastography imaging system is composed of two parts:
excitation and detection. Compared with conventional ultrasound elastography [68], high
frequency ultrasound elastography [22, 23] and magnetic resonance elastography [69],
utilizing optical coherence elastography (OCE) as the detection method, have gained the
36

capability to characterize subtle stiffness changes in ocular tissue because of the high
resolution (<10 µm) and the advantage in transparency [29, 70].
To induce a small temporal axial displacement to launch a mechanical shear wave in
ocular tissue, multiple excitation methods have been explored, including air-puff [71, 72],
acoustic micro-tapping [55], and acoustic radiation force (ARF) [73-75]. To be specific,
current air-puff OCE was focused on producing mechanical waves in the cornea and has
not been applied on the posterior eye. In addition, air-puff pulse suffers from few
limitations, including the low bandwidth of the induced mechanical waves and slow
relaxation times [29]. Another excitation approach is so-called acoustic micro-tapping. In
general, a spatially and temporally sharp pressure is applied to the tissue surface via a
focused air-coupled ultrasound transducer. Because a major part of the acoustic intensity
is reflected at the boundary, only less than 1% intensity was applied to the tissue, launching
a wave with nano to micrometer displacement. Owing to the stiff sclera tissue and large
attenuation from the long propagation distance to posterior eye, the feasibility of applying
this approach to characterize the elasticity of posterior eye still needs further investigation.
In order to have a more controllable localized force inside the imaging region, Qu et al.
[57] first applied ARF-OCE technique to characterize the biomechanical properties of the
retina in vivo and successfully reconstructed the elasticity of each retinal layer. However,
owing to the large acoustic attenuation of the anterior chamber, especially of the lens, a
high power ARF is typically desired to induce sufficient deformation at the posterior eye
directly. As a consequence, its mechanical index (MI) and acoustic intensity remain a
challenge to meet the strict U.S Food and Drug Administration (FDA) standards for
ophthalmic exposure.
In this study, we report on the development of a shaker-based OCE technique as a
potential tool for clinical study. The proposed method has the capability to assess the
elasticity of the posterior eye, including different layers of the retina via shear wave
elastography. The performance of the system has been validated on both a homogeneous
phantom and a healthy rabbit eye in vivo.

37

4.2 Methods
4.2.1 Experimental setup
A schematic diagram of the experimental setup is shown in Figure 4-1. To induce the
shear wave, a mechanical shaker (mini-shaker type 4810; Bruel & Kjaer, Duluth, Georgia,
USA) was positioned at the anterior sclera which is close to corneal limbus via a contact
rod. The tip of the rod has a square shape with the size of 1.2 mm by 1.2 mm. Then, the
shaker was aligned along the scanning direction of the OCT beam.

Figure 4-1. Shaker-based OCE system schematic diagram with in vivo rabbit eye setup. SLD:
superluminescent diode: OC: optical coupler: CO: collimator: OA: optical attenuator: M: mirror: GM:
galvanometer mirrors: L1/L2: lens: RFA: radiofrequency amplifier: FG: function generator: G: grating.

To precisely track the tissue motion caused by mechanical shaker, a customized
50 kHz spectral domain optical coherence tomography (SD-OCT) system with a central
wavelength of 890 nm and bandwidth of 144 nm was built. For the safety purpose, the
light beam was filtered through an optical isolator (IO-F-SLD150-895, Thorlabs Inc.,
Newton, NJ, USA) and then split 20% to the imaging sample and 80% to a reference
mirror using an optical coupler. Glass imaging windows are placed in the stationary
reference arm for dispersion compensation. During the experiment, the scattering signal
38

from the sample arm is first coupled together with the reflected reference arm signal, and
then generates the interference signal. Next, the signal is separated by wavelength with a
diffraction grating and focused onto a line scan CMOS camera. Finally, the raw signals
were saved offline for further processing. For the purpose of the shear wave tracking, the
shaker was fixed, while galvo mirrors scan at M-B mode with a step size of 6 µm. A total
of 3 mm scanning distance was acquired.
To synchronize the shaker and the SD-OCT detection system, the PC sent out a
baseline signal to trigger the arbitrary function generator (AFG 3252 C, Tektronix,
Beaverton, OR, USA), which generates an impulse signal with a pulse width from 200 µ s
to 1 ms. Then the impulse signal was transmitted to a power amplifier (Type 2718, Bruel
& Kjaer, Duluth, Georgia, USA) to generate an amplified impulse signal to the
mechanical shaker to induce tissue deformation. At each lateral scanning location, M-
mode collected a total of 400 A-lines, corresponding to the time of 8.8 ms, before the
galvanometer moves to the next scanning position. To establish the baseline for the
displacement curve, the mechanical shaker was excited to 100 µ s after the SD-OCT
system started to acquire data. These parameter settings were kept constant for both
phantom study and in vivo animal study.

4.2.2 Post-processing and data analysis
Data analysis was performed using MATLAB. The signal is processed and transformed
into depth-resolved intensity and phase-resolved displacement. To calculate the shear wave
speed (SWS), the spatiotemporal map (lateral distance versus time shifts curve) was
obtained from the axial displacement map where the lateral distance was measured by the
moving step size of the galvo mirror and the time shift (defined as the time to reach the
peak displacement at each dynamic displacement). Then, the SWS was estimated by
applying a linear regression to all peak displacement points along successive lateral
locations. To quantify the Young’s modulus, the Eq-2 was used. The measurement was
repeated for at least three times to achieve the statistical significance.

39

4.2.3 Phantoms and rabbit preparation
A custom-built silicone phantom was first fabricated to test the performance of the
system. The stiffness of the homogeneous phantom was tested by the gold-standard –
uniaxial mechanical testing (Model 5942, Instron Corp., MA, USA), which is equal to
73.9 ± 7 kPa.
The in vivo rabbit experiment was performed according to the University of Southern
California Institutional Animal Care and Use Committee (IACUC) protocol. Dutch belted
pigmented rabbit (∼2 kg) was single housed and fed with normal diet. Prior to the imaging
experiment, the rabbit was anesthetized with ketamine (35 mg/kg) and xylazine (5 mg/kg)
via subcutaneous injection. Two drops of phenylephrine were applied topically to prevent
cornea swelling and decrease discomfort. Redosing of anesthesia was achieved by 2.5%
sevoflurane through a facial mask. The heart rate, respiratory, body temperature, oxygen
saturation, and non-invasive blood pressure were recorded every 5 min until it became fully
conscious.

4.3 Results
4.3.1 Phantom results
The shaker was positioned at 4.8 mm laterally away from the OCT detection beam in
order to mimic the potential shear wave propagation distance from anterior sclera to
posterior segment of the eye. Figure 4-2. (a) shows the displacement curves under the 200
µ s, 600 µ s, and 1000 µ s shaker duration which was acquired at the initial galvo position,
respectively. It was observed that longer duration provides better signal-to-noise (SNR)
ratio. However, for the purpose of safety and maintaining a wide bandwidth of the
generated shear wave, 600 µ s was selected as a balance and further implemented for the in
vivo rabbit study. Figure 4-2 (b) shows the displacement curves at three different depths of
the phantom. It was observed that the peak displacement and the time to reach the peak
displacement are similar among three depths, indicating the uniform force distribution in
axial direction.
40

Figure 4-2. Displacement curves of the homogeneous phantom. (a) With 200 µ s, 600 µ s, and 1000 µ s shaker
pulse duration at the initial galvo position (the most left position of the OCT image). (b) At the depth of
50 µm, 250 µm, and 450 µm under the condition of 600 µs shaker pulse duration. (A color version of this
figure is available in the online journal.)

Figure 4-3 (a) to (c) shows the OCT image, spatiotemporal map, and shear wave
propagation at three different time points, respectively. The effective field of view (FOV)
of shaker-based OCE system can reach up to 600 µm, and therefore it is sufficient to cover
the imaging subject where the healthy rabbit retinal thickness averages about 300–400 µ m.
To assess the accuracy of our proposed shaker-based OCE method on estimating Young’s
modulus, we compared it with the gold standard – uniaxial mechanical testing (Instron
5942). Calculated from the spatiotemporal displacement map of the phantom, the SWS
determined by the linear regression of the ratio between the propagation distance and the
time to reach the peak displacement in successive lateral locations is 5.1 ± 0.3 m/s,
corresponding to the Young’s modulus of 78.3 ± 9 kPa. The calculated Young’s modulus
is consistent with 73.9 ± 7 kPa tested on gold standard. The difference between the
reconstructed Young’s modulus via SWS and the gold standard is not significant (P
value = 0.4133, where P < 0.05 is considered to be significantly different), and is within the
acceptable range of error (∼5%).
41

Figure 4-3. (a) OCT image of the phantom, (b) spatiotemporal displacement map of the homogeneous
phantom, (c) shear wave propagation at the timing of 0, 0.088, 0.176 ms, respectively.

It was also observed in Figure 4-2 that the first 1 ms in axial displacement curve at the
initial galvo position has no displacement, which corresponds to the propagation time
from the location of shaker to the region of interest (ROI) of the phantom plus the
synchronization delay between shaker and SD-OCT system. By removing the
synchronization delay time, we got the pure shear wave propagation time of 900 µ s. As a
result, the average SWS in this region was estimated to be 5.3 m/s, which is close to the
calculated 5.1 m/s inside the OCT beam scanning region. All these results demonstrated
that our imaging method has the capability to assess the Young’s modulus precisely.

4.3.2 In vivo rabbit posterior eye
Imaging was performed on the central retina and the same M-B mode scanning scheme
was used to capture the shear wave propagation. The structural resliced OCT images were
42

obtained as shown in Figure 4-4 (a), where the individual five posterior layers of the eye
could be isolated. Figure 4-4 (b) to (d) shows the shear wave propagation at time of 0,
0.088, and 0.176 ms, respectively. In Figure 4-5 (a) to (d), four spatiotemporal
displacement maps at layers 1, 2, 3, and 5 (indicated in the Figure 4-4 (a)) are plotted,
respectively. The SWS of the first three layers from the ganglion side to the photoreceptor
side are: 4.1 m/s, 4.9 m/s, and 6.7 m/s, which are corresponding to the elasticity of 50.4 kPa,
72 kPa, and 134.6 kPa, respectively. Owing to the low OCT signal in layer 4, the SWS of
the choroid was not identified here. The layer 5 – sclera has an average SWS of 9.1 m/s
which corresponds to the elasticity of 248 kPa.

Figure 4-4. (a) OCT image of the posterior rabbit eye in vivo, (b–d) shear wave propagation maps at the
timing of 0, 0.088, 0.176 ms, respectively. Layer 1: Nerve fiber, ganglion cell, and inner plexiform; Layer 2:
inner nuclear, outer plexiform, and outer nuclear; Layer 3: RPE; Layer 4: choroid; Layer 5: sclera.

Figure 4-5. Spatiotemporal displacement maps of different layers of the posterior eye. (a) Layer 1, (b) Layer
2, (c) Layer 3, and (d) Layer 5.

43

4.4 Discussion and Conclusion
Optical coherence tomography is a well-developed imaging technique to provide
excellent spatial resolution and can be used to characterize the tissue biomechanical
properties when combined with an external inducing force. In recent years, quantified
elasticity maps of the in vivo posterior eye are first presented by using ARF-OCE technique.
However, ARF-based technique suffers from two concerns. One is non-uniform ARF beam
and relative shallow ARF excitation region in the axial direction; another concern is the
safety issue. The mechanical index (MI) for ocular applications is 0.23 as determined by
the FDA. In order to maintain a high SNR, a large output power of ARF is typically
required, which impedes its translation to clinical studies. By contrast, shaker-based
method has been previously investigated by ultrasound elastography studies on human
subjects in clinic, including cornea [76] and posterior sclera [5], which may be a relatively
safe approach.
In this study, we used a shaker to mechanically induce the tissue deformation. To be
specific, the force is applied orthogonally to the anterior sclera surface in order to provide
shear wave propagation only [28]. As indicated by Palmeri et al. [77], the frequency
bandwidth of phase velocity depends on the excitation duration and spatial beamwidth used
to generate the shear wave. Excitation durations from 100 to 700 µs (less than 1 ms) are
typically used to induce shear wave propagation [78]. Based on these previous studies, we
concluded that shorter excitation duration can obtain wider shear wave bandwidth,
resulting in more accurate estimation of elasticity. However, previous ARF-OCE system
utilized a relative long ARF excitation duration. This is because that a short excitation
duration (below 1 ms) would not induce enough detectable displacement at posterior eye,
especially when the shear wave propagates far away from the region of excitation of the
pushing transducer [21, 57]. Compared with ARF method, the minimally contact shaker
method is capable of providing a shorter pushing duration – 600 µ s is used in this study,
while maintaining sufficient tissue deformation.
The performance of the shaker based OCE system was first validated on a homogeneous
phantom. The variance of the peak displacements among different axial depths in Figure
4-2 (b) is less than 5%, which implies that the shaker has the potential to generate uniform
force distribution within the OCT image view. In addition, the Young’s modulus of the
44

phantom reconstructed by the linear regression of the peak displacement points in the
spatiotemporal map is closely relative to the mechanical testing results. All these results
demonstrated that the shaker based OCE has the ability to accurately capture the shear
wave propagation with a large field of view. To further validate the potential capability of
our system on pre-clinical in vivo study, the posterior rabbit eye was imaged.
There are few literature studies on the elasticity range of the posterior eye. Chen et al.
[79] identified the elastic modulus of the retina ex vivo using mechanical testing. However,
this test is performed on porcine tissue and without perturbing the natural retinal
environment. Qu et al. [57] first demonstrated the mechanical quantification of the in vivo
posterior eye using the bulk frequency response method. In a healthy rabbit model, their
mechanical properties vary from 3 to 16 kPa in different layers of the retina. However, the
measured biomechanical properties depend on many parameters such as test conditions,
species, and most importantly, measurement technique. The resonance frequency-based
elasticity measurement is highly dependent on the characterization of the tissue and the
implemented model. Later, He et al. [21] reconstructed the elasticity of the retina using
shear wave OCE. In their report, the elasticity of the first three layers from the ganglion
side to the photoreceptor side ranges from 12 kPa to 101 kPa. In addition, they indicated
that the Young’s modulus of the bottom two layers is over 100 kPa. Our calculated Young’s
modulus from anterior retina to posterior retina and sclera are in a reasonable range
compared with results above.
In this study, we successfully demonstrate the capability of our imaging system to
characterize the biomechanical properties of the posterior eye. Considering the high safety
requirements in clinical study, our shaker-based OCE system may have the translational
potential for clinical diagnosis. However, a few challenges remain to be addressed before
the technology can be translated. First, the group velocity-based elasticity assessment in
this study maybe inaccurate (i.e. under the assumption of pure elasticity and homogeneous
medium) and may lead to some bias because of boundary conditions and the ratio of retina
thickness to shear wavelength. A multi-layer model (modified from previously developed
Lamb wave model [28]) is needed to precisely reconstruct both elasticity and viscosity of
different retinal layers. Second, as suggested by Kirby et al. [29], the bandwidth of
generated mechanical waves depends on the temporal and spatial characteristics of the
45

excitation push. Although 600 µ s is a short push, the push cross section determined by the
size of shaker’s contact rod is not tiny in this study, which may reduce the mechanical wave
bandwidth. They also pointed out that the bandwidth/wavelength relationship of the
mechanical wave determines the spatial resolution of reconstructed elastic modulus maps.
In the future study, a tiny rod will be implemented, which may help us to further
characterize the reconstructed elastic modulus maps. Finally, the safety issue of the shaker-
induced process still needs to be further investigated to ensure its potential applications in
clinical settings.
In summary, we demonstrate that the shaker-based OCE method has the ability to
accurately map the biomechanical properties of different retinal layer and sclera underneath.
Compared with ARF-OCE approach, our developed method can induce sufficient shear
wave propagation at the posterior eye with high resolution and large field of view, which
could lead to a quicker clinical uptake.
46

5. Chapter 5 High Resolution Optical Coherence Elastography
of Retina under Prosthetic Electrode
5.1 Introduction
Diseases of the posterior segment of the eye, including age-related macular degeneration
(AMD) and retinitis pigmentosa (RP), damage photoreceptors and their supporting
epithelial cells [80]. More than 1 million patients are affected by RP worldwide. Initially,
patients retain their central visual field with gradual peripheral vision loss, however,
eventually they will go on to lose both areas of vision with progression of the disease [81].
AMD is the leading cause of blindness in patients aged 65 or older in developed countries,
with more than 8 million Americans suffering from the disease [82]. Currently, there is no
known cure to reverse the progressive loss of photoreceptors due to these two conditions.
Retinal prosthetic electrodes hold the potential to restore some visual functions by
directly stimulating the neural retina with electrical pulses [83-86]. The Argus® II retinal
prosthetic electrode (SSMP, Sylmar, CA, USA) is the first and only commercial product
that has received the European conformity mark and been approved by the U.S. Food and
Drug Administration (FDA) [87-89]. With Argus II, a camera mounted to a pair of glasses
worn by the patient first captures the visual signals from the patient’s environment and
these are then sent to a video processing unit, through a transmitting coil. The received
electrical signals are then used to drive corresponding in-eye mounted electrodes which
then directly stimulate the patient’s retina thereby creating the experience of vision where
previously none existed [90-93]. A flexible polyimide (PI) based microelectrode array
stimulates the epiretinal side of the retina. The safety and bio-compatibility of PI electrodes
has been previously investigated, however, the biomechanical effects of retinal prosthetic
electrodes on the posterior segment of the eye still remain unknown [94-97].
Biomechanical properties of ocular tissues are crucial for the health of both the anterior
and posterior segments of the eye, with many ocular diseases being associated with altered
biomechanics. Keratoconus is a prevalent disease which leads to significant visual
impairment due to the development of a cone-shaped ecstatic cornea, and one of its clinical
signs is an unusually compliant cornea compared with normal cornea [1]. It has been shown
that mechanical properties of the posterior segment of the eye change with progression of
47

diseases such as glaucoma [8]. The biomechanics of lamina cribrosa (LC) has been
postulated to play a pivotal role in ganglion cell dystrophies in glaucoma [10, 14]. The
retina also suffers mechanical alteration with progression of blood vessel infiltration due
to AMD [21].
Elastography is a widely used imaging modality to reconstruct biomechanics in soft
tissue in a non-invasive manner, and is often used to provide supplementary diagnoses for
standard structural imaging. There are several methods available for measuring
biomechanics by elastography, including conventional ultrasound elastography [5, 98],
high resolution ultrasound elastography [22, 27, 28] and optical coherence tomography
(OCT) based elastography (OCE) [99, 100]. Among these techniques, OCE is the most
promising imaging modality in the field of ophthalmology owing to the advantages of high
resolution (<10 µ m), high sensitivity, and the transparency of ocular tissue. OCE has been
widely used in mapping the elasticity of ocular tissue, such as cornea [19], lens [20], retina
[30], and optic nerve head [33].
To initiate the propagation of an elastic wave, several excitation methods have been
employed, such as air-puff [72], acoustic micro-tapping [55], and acoustic radiation force
(ARF) [19, 33]. Both air-puff and acoustic micro-tapping are capable of reconstructing
biomechanical properties of the anterior segment of the eye. However, due to the natural
geometry of the eye and attenuation that occurs over the long distance from the anterior to
the posterior segment, these methods are not optimal in this study. Qu et al. first utilized
ARF-OCE to reconstruct the biomechanical properties in each layer of retina in vivo [19].
To induce desirable tissue motion to track elastic waves, high power ARF is required. Due
to strict FDA regulation in ophthalmic exposure to ultrasound, the mechanical index and
acoustic intensity prevent this technology from being used in the clinic. Zhou et al. utilized
a shaker to induce harmonic vibration, and an ultrasonic array to measure induced wave
propagation in the posterior sclera in humans [101]. Qian et al. applied a shaker as a
vibrating source to induce tissue motion, and demonstrated feasibility for reconstruction of
biomechanical properties of the retina in rabbit eye [30].
In this study, we propose to use a shaker-based OCE technology to evaluate
biomechanical properties in the retina with prosthetic electrodes, potentially supplementing
48

the data set used for evaluation of the Argus II retinal prosthetic electrode, and then
translating this technology into clinical use.

5.2 Methods
5.2.1 Experimental setup
System settings have been shown elsewhere [30]. Briefly, as shown in Figure 4-1, a
spectral domain optical coherence tomography system (SD-OCT) with center wavelength
of 890 nm and bandwidth of 144 nm was developed to image tissue structure as well as to
track elastic wave propagation. A mechanical shaker (mini-shaker type 4810; Bruel &
Kjaer, Duluth, Georgia, USA) was used to induce tissue motion and initiate elastic waves.
The rod tip of the shaker was placed on the corneal limbus and aligned in the direction of
the OCT scanning beam with real time Doppler OCT imaging. A step size of 6 µ m was
used for the scanning galvanometer mirror positioning system (galvo) and a total of 500
positions were scanned.
To precisely induce and track elastic wave propagation, all components of the system
were synchronized. A baseband signal generated by the personal computer (PC) triggered
an arbitrary function generator (AFG 3252C, Tektronix, Beaverton, OR, USA). A 0.6 ms
pulse-width impulse signal was used to generate detectable axial displacement with broad
bandwidth [30]. The impulse signal was amplified using a power amplifier (Type 2718,
Bruel & Kjaer, Duluth, Georgia, USA) and finally transmitted to the shaker. A series of
400 A-lines was acquired at every position corresponding to 8.8 ms before the galvo
translated to the next position. Inter-A-line analysis was performed with a 20 µ s time
interval to obtain axial displacements for further processing [20]. Inter-A-line analysis was
insensitive to motion artifacts because the data acquisition frequency was much faster than
the motion frequency, and adequate anesthesia was administrated to reduce the motion
artifact. The optical table and sound absorption sponge were also used to reduce the
unanticipated vibrations from the environment.

49

5.2.2 Post-processing and data analysis
After scanning in each lateral position, raw data was saved to disk for off-line processing.
Depth-resolved OCT intensity data and phase-resolved Doppler OCT data were first
obtained. To calculate the elastic wave speed, the spatial-temporal map was reconstructed
at each lateral position and time. The elastic wave speed can be estimated with known
propagation distance and time. To precisely calculate the wave speed, linear regression was
used for all successive peak axial displacements at each lateral position in the
corresponding spatial-temporal map. In this study, the elastic wave speed in each layer was
measured before and after the implant of prosthetic electrode, the Shapiro-Wilk test was
performed to evaluate the normality of the speed distribution, and the paired t-test was
performed to evaluate the statistical significance.

5.2.3 Implant surgery
Surgery protocol was approved by the University of Southern California Institutional
Animal Care and Use Committee (IACUC). Prior to surgery, Dutch Belted Pigmented
rabbits (~2 kg) were anesthetized with ketamine (35 mg/kg) and xylazine (5 mg/kg)
through subcutaneous injection. Additional anesthesia was induced using 2.5% sevoflurane
through a facial mask. One drop of 0.5% proparacaine hydrochloride ophthalmic solution
was administrated topically for ocular anesthesia, and 1 drop of 1% tropicamide and 2.5%
phenylephrine was applied in the same way for eye dilation. Heart-rate, respiration, oxygen
saturation, body temperature, non-invasive blood pressure, oxygen flow and end tidal
carbon dioxide (when sevoflurane was used), were recorded every 5 minutes while the
animals were unconscious. First, phacoemulsification was performed for all three animals,
and the OCE imaging was conducted for 3 times in each animal. Intra-vitreous injection of
0.3 mL sterile air was performed and then the eyes were treated with antibiotic ointment
and 1% atropine immediately to prevent infection and to maintain postoperative
cycloplegia. One week later, a 3-port 23-G pars plana vitrectomy (PPV) was performed
with the Stellaris PC platform (Bausch & Lomb, Rochester, NY, USA) in a sterile
environment. A microelectrode was then inserted through a temporal sclerotomy
(approximately 4 mm in width) and was placed onto the retina below the optic nerve head
50

(ONH) and then flattened using perfluorocarbon liquid to keep it securely attached to the
retina after vitrectomy. A tag was usually used in clinical trial to immobilize the prosthesis
onto the retina, however, considering the difference in the thickness of the ocular tissue
between human and rabbit, the tag has not been used to avoid failure of the surgery. The
extraocular portion of the cable was sutured to the sclera below the conjunctiva. At the end
of the procedure, the sclerotomies and conjunctiva were sutured. After the operation,
topical antibiotic eye drops were used 4 times daily. And OCE imaging was performed 3
times in each animal with the implant. No complications related to surgery were observed
during or after surgery.

5.3 Results
5.3.1 Prosthetic electrode characterization
The requirements for retinal prosthetic electrodes mainly include high electrode density,
biocompatibility, and flexibility [102, 103]. A high-density prosthetic electrode allows
high resolution stimulation and hence improves visual acuity. Due to the natural curvature
of retina, flexibility enables the prosthetic electrode to adhere to the surface of the retina
seamlessly for a prolonged period of time. Polyimide has the advantages of flexibility,
mechanical stability, biocompatibility, and thermal stability, and was therefore used as the
structural material in the prosthetic electrode. The three-dimensional platinum (Pt)-pillar
coating method was used to increase the electrode density and constrain the electrode size
while maintaining flexibility of the entire structure [104].

Figure 5-1 a shows the structure of the prosthetic electrode. The multi-electrode array is
connected with a high lead count cable for retinal stimulation. The array has 14× 9 elements
with a diameter of 150 µ m each. The length of the prosthetic electrodes and the array is 37
and 3.5 mm, respectively. Figure 5-1 (b-d) demonstrates the prosthetic electrode’s
flexibility which facilitates seamless bonding to the complex surface of the retina.

51

Figure 5-1. Characterizations of retinal prosthetic electrode. (A) Structural image of the prosthesis; (B,C,D)
optical images of the flexible prosthesis when it stands freely, bends convexly and concavely with more than
90° .

5.3.2 Surgery characterization
Post-surgery evaluation is essential in this study. Failure of implantation is common in
phacoemulsification and implantation procedures. Hemorrhage and iris bleeding are also
rarely observed during vitrectomy due to the high viscosity of vitreous humor [105, 106].
Retinal detachment can occur during prosthetic electrode implantation when vitreous
humor is not fully removed prior to the procedure. Figure 5-2 demonstrates successful
implantation and bonding of the prosthetic electrode to the retina. Figure 5-2 (a-c) show
cross-sectional OCT images with the prosthetic electrode. Figure 5-2 (d-e) show the en
face RetCam (Natus Medical Incorporated, Pleasanton, CA, USA) images after
phacoemulsification and implantation respectively.
52

Figure 5-2. a-c Cross-sectional OCT images of retina with prosthetic electrode. d-e RetCam images of retina
after phacoemulsification procedure and implantation procedure respectively. ONH: optic nerve head, RV:
retina vessels.

5.3.2 Biomechanical response of retina to the prosthetic electrode
To investigate the influence of the prosthetic electrode on the retina, OCE imaging
was performed after phacoemulsification and implantation. As with our previous work
[30], here we segment the raw measured data to five different layers according to the
anatomy of the retina and assign them respective numerals i through v for efficient
classification. Layer i to layer v correspond to the nerve fiber, ganglion cell, and inner
plexiform layer; inner nuclear, outer plexiform, and outer nuclear layer; retinal pigmented
53

epithelium layer; choroid layer; choroid and sclera layer respectively. Spatial-temporal
maps were constructed in each layer along with lateral positions and time except for layer
iv due to its low OCT signal which is similar to our previously reported study [30].
Figure 5-3 shows the spatial-temporal map in each layer with and without the prosthetic
electrode. The slope of the wavefront is the elastic wave speed in the corresponding
retinal layer. Figure 5-4 shows the statistical analysis of elastic wave speed in each layer
with and without the prosthetic electrode; the error bar represents the standard deviation.
Without the prosthetic electrode, the elastic wave speed of layers i, ii, iii, and v is
3.66± 0.36, 5.33± 0.07, 6.85± 0.37, and 9.69± 0.24 m/s, respectively. With the prosthetic
electrode, the elastic wave speed in each corresponding layer is 4.09± 0.26, 5.14± 0.11,
6.88± 0.70, and 9.99± 0.73 m/s, respectively. The difference between the results in each
group and each layer were evaluated using a Shapiro-Wilk test and a paired t-test. The
statistical analysis showed that the elastic wave speed distributes normally with P<0.05 in
each layer of the retina. The elastic wave speed in retina with electrode was slightly
higher where P<0.05 with an average of 2.03 m/s, and the 95% confidence interval was
from 0.53 to 2.82 m/s. The elastic wave propagates faster along the depth direction of the
retina, indicating the elasticity increased from the ganglion side to the photoreceptor side.
The photoreceptor side is close to the sclera. The sclera is stiffer than retina, and this
close-knit structure of retinal layers is likely to affect the biomechanical properties.

54

Figure 5-3. Spatial-temporal maps in each layer of retina with and without prosthetic electrode. The slope of
the white line represents the corresponding elastic wave speed. Color bar represents the axial displacement.

Figure 5-4. Statistical analysis of each layer in the retina with and without prosthetic electrode.

55

5.4 Discussion and Conclusion
In this paper, we have presented the first reported evaluation of biomechanical effects
at the back of the eye in the presence of retinal prosthetic electrodes. We developed an
OCE system with a mechanical shaker to induce and track elastic wave propagation in the
posterior of the eye. Our highly sensitive and high spatial resolution OCE system enables
detection of axial displacements precisely, which leads to high accuracy of subsequent
post-processing and wave speed estimation. We used a mechanical shaker as a vibrating
source to induce tissue motion. Although the shaker we used currently cannot be confocal
with the OCT scanning beam, it has great practical potential for translation to clinical use
compared with other vibration sources. The current tilted shaker configuration may induce
complex waves instead of pure shear waves which may lead to some bias in the estimation
of the wave speed. However, in previous work we showed the accuracy of our system with
similar settings to those that were used here [30]. The reconstructed Young’s modulus of
imaged phantoms was comparable to uniaxial mechanical testing with a stand-alone
elasticity measurement system (Model 5942, Instron Corp., MA, USA).
Previous work to evaluate the performance of the prosthetic electrode included
perceptual threshold and electrode impedance calculation, visual functions testing, and
analysis of the morphology of both the retina and prosthetic electrode [94, 95]. The tests
confirmed the safety and reliability of the prosthetic electrode but did not investigate
biomechanical properties of the eye in the presence of the implanted electrode. In the
current study, we investigated how the prosthetic electrode affects the biomechanical
properties of the retina to help further knowledge and general understanding of these
important effects.
In this study, we measured the elastic wave speed before and after prosthetic electrode
implantation. Results showed that the measured biomechanical properties of the retina are
comparable in each layer of tissue with and without the prosthetic electrode, and also
remained consistent with biomechanics of the unaltered retina as reported in previous work
[30]. The prosthetic electrode is mainly composed of PI which has a density of ~1.4 g/cm3
and is similar to the density of fluid in the posterior chamber of the eye. Furthermore, the
flexibility of the lightweight prosthetic electrode allowed it to bond seamlessly to the retinal
56

surface, thereby eliminating unwanted strain force which limited the biomechanical effects
on the retina.
Although we have successfully demonstrated the effects of the prosthetic electrode on
retinal biomechanical properties, there are still some challenges that can be addressed in
future work. First, elastic wave speed was used for evaluating the biomechanics. Previous
studies have shown that this method can induce bias estimation in thin, viscous, anisotropic,
and inhomogeneous media [29]. To solve this problem, Shih et al. [28] and Han et al. [107]
implemented a Lamb Wave model using ultrasound elastography and OCE respectively to
calculate the elasticity and viscosity of the cornea. The Lamb Wave model assumes the
imaged target is a single thin layer, however, the retina has 5 layers, making the model
inappropriate in this study. Zvietcovich et al. [108] developed reverberant 3D OCE to
capture the elastic wave propagation in each layer of the cornea with excellent contrast. In
future work, we will implement an appropriate model for biomechanical property
reconstruction of the retina. Second, the quality of our OCT results is not optimal compared
with OCT images of the retina in normal eyes. One important reason for this observed
effect was that the eyes we used underwent phacoemulsification, vitrectomy and
implantation. All of these procedures were likely to cause corneal swelling due to
intraocular pressure (IOP) variation when balanced salt solution (BSS) was infused into
the eye. The cornea therefore experienced excess stress and strain, which damaged the
endothelium cells and led to blurring. A second important cause of reduced OCT quality
in these studies is that intraocular inflammation usually occurs during vitrectomy [105,
106], resulting in intraocular blurring. Last, the metal electrode blocked the optical path of
the OCT system, with the beam only able to penetrate through the gap between electrodes
which led to a decrease in the OCT signal. All of these factors degraded the OCT image
quality in this study. While these issues are currently difficult to resolve with our current
system and setup, in future work we plan to use improved surgical techniques and a new
high power OCT system to improve the image quality. This study is also limited by the
lack of electrical stimulation during OCE imaging. However, the aim of this study was to
explore the effects of the prosthetic electrode on the biomechanics of the retina. Further
studies are needed to apply a range of electrical stimuli and investigate the corresponding
biomechanical effects.
57

In summary, our work demonstrated surgical implantation of a prosthetic electrode in
rabbit eye, and the biomechanical response of the retina to the prosthetic electrode. Based
on the elastic wave speed from the measured data in each layer of the retina with and
without the prosthetic electrode, we conclude that the prosthetic electrode does not affect
the biomechanical properties of the retina significantly. We hope that this technology can
be further translated into the clinical use so that it can evaluate retinal biomechanics in
patients with retinal prosthetic electrodes for longitudinal study over an extended period of
time.

63

6. Chapter 6 Simultaneous Assessment of the Whole Eye
Biomechanics Using Ultrasonic Elastography
6.1 Introduction
The eye is a complex organ that senses light and encodes this information into neural
stimuli to form images in the visual cortex [19]. Each component of the eye plays an
important role in maintaining normal visual function. The cornea and the crystalline lens
primarily focus light into the posterior segment of the eye based on their transparency,
curvature, and morphology. Around 60% and 40% of the focusing power of the eye are
attributed to the cornea and lens, respectively [20]. The iris is the colored part adjacent to
the lens, which controls amount of light entering the posterior segment to accommodate
changes in environmental light intensity [109]. The posterior segment of the eye includes
the neurosensory retina, which receives the light signal [110] and transmits a neuronal
signal to the brain via the optic nerve. Dynamic force loads on the optic nerve head (ONH)
are supported by the lamina cribrosa and peripapillary sclera (PPS) [111, 112].
Despite the complexity of their individual functions, some properties of ocular tissues
are inter-related. Ocular diseases can have global effects on both morphological and
biomechanical properties of the whole. One example is primary open-angle glaucoma
(POAG). POAG is a leading cause of irreversible blindness worldwide, and there is
currently no cure for the disease [2]. High intraocular pressure (IOP) is a primary risk factor
for glaucoma; high IOP increases the tension force on the ONH, thereby damaging the
optic nerve. Vision loss gradually develops as the nerve degenerates. However, the ONH
is not the only ocular tissue to its biomechanical properties altered by elevated IOP; other
ocular tissues, including the cornea, are also affected [113].
Current ophthalmic diagnostic tools mainly include the slit lamp, optical coherence
tomography (OCT), ocular photography, and ultrasonic biomicroscopy (UBM). These
tools support visualization of anatomical structures to detect abnormal morphologic change.
However, biomechanical alterations are also vital signs to forecast or diagnose some ocular
diseases [114]. For example, reduced corneal stiffness is an most risk factor for keratoconus
64

[115]. Therefore, a diagnostic tool that can directly assess or detect these biomechanical
alterations could be beneficial for scientific research and clinical care.
Elastography is an emerging imaging modality that can provide measurements of tissue
elasticity in a non-invasive manner. Ultrasonic elastography (UE) and optical coherence
elastography (OCE) are two imaging modalities that are applicable in the field of
ophthalmology. These imaging modalities have their own advantages and limitations; OCE
provides better spatial resolution but worse imaging depth compared to UE. OCE has been
widely used in ophthalmology research owing to the transparency of several ocular tissues
and the high resolution of this imaging modality [30]. In ophthalmology, the research of
OCE mainly focus on the anterior segment. Singh et al. conducted compressional OCE to
measure corneal elasticity before and after crosslinking [116]. Zvietcovich et al. performed
3D-reverberant wave OCE to assess the elastic wave distribution through each corneal
layer with enhanced contrast [108]. Yan et al. quantified the biomechanical properties of
both the cornea and crystalline lens using OCE by integrating a swept light source to
increase the imaging range [20]. Although the imaging depth of OCE has been significantly
improved through these modifications, OCE is still unable to image the whole eye at once.
There have been significant progresses in the application of UE in the field of
ophthalmology. Kwok et al. used a 50 MHz linear array to measure induced corneal axial
displacement induced by heartbeats. While this technology does not require an external
vibrator, it produces only strain imaging [117]. Zhou et al. used a vibrator as the pushing
source and a 6.4 MHz linear array to capture the elastic wave of sclera in vivo in human
eyes; while the low frequency of ultrasonic array increased imaging depth, imaging
resolution was decreased [5]. Weng et al. combined high frequency (40 MHz) UE to
quantify the biomechanics of the cornea [118]. This approach provides good imaging
resolution and image quality; however, the imaging depth is limited due to the high
attenuation of high frequency ultrasound. Therefore, any approach requires a compromise
between imaging depth and resolution; it is important to choose an appropriate frequency
range for the imaging transducer array to achieve appropriate imaging depth and resolution.
In this study, we propose to develop a UE system that uses a mechanical shaker as the
vibration source and a moderate frequency range (18 MHz) linear array to provide deep
imaging depth with good resolution. To the best of our knowledge, this is the first imaging
65

method that supports mapping and quantification of the biomechanical properties of ocular
tissues of the whole eye simultaneously, including the cornea, lens, iris, ONH and PPS.

6.2 Methods
6.2.1 Experimental Set-up
The experimental set-up is shown in Figure 6-1. In both the phantom validation study
and in vivo animal study, the same parameters and set-up were used. The mechanical
shaker was driven by a function generator (AFG 3252C, Tektronix, Beaverton, OR, USA)
and a power amplifier (Type 2718, Bruel & Kjaer, Duluth, Georgia, USA) to initiate gentle
stimuli for elastic wave propagation. A programmable ultrasound research platform
(Vantage 256, Verasonics, Redmond, WA, USA) and an 18 MHz 128 element linear array
(L22-14v, Verasonics, Kirkland, WA, USA) was used for B-mode imaging and ultrafast
elastography. A rod with a diameter of 0.5 mm was used to indent the tissues. The rod was
connected to the shaker and aligned in the same plane with the array and touched either the
phantom or sclera perpendicularly. The diameter of the rabbit eye is typically less than 17
mm; therefore, an imaging depth of 20 mm was selected for this study.

Figure 6-1. Schematic diagram of the shaker based ultrasonic elastography system for the in vivo study.
66

An IOP control and monitoring system was established to manipulate the IOP for the in
vivo experiment as shown in Figure 6-2. The 23 G trocars were inserted 1.5 mm away from
the limbus to establish ports to access the intraocular contents. A pressure sensor was
inserted into one port to monitor IOP changes in real time. An infusion line that contained
aseptic balanced salt solution (BSS) was inserted into the other port. IOP was controlled
by modulating the heights of the infusion bottle and line.

Figure 6-2. Schematic diagram of the IOP control and monitoring system.

6.2.2 Data Processing
The demodulated in-phase/quadrature data was saved for off-line processing after each
scan. The axial displacement was assessed using a 1D autocorrelation algorithm [119]. A
3D data matrix was acquired that contained data on lateral position, depth position, and
time with the axial displacement as the index. The spatial-temporal map was reconstructed
at each lateral position and time to calculate the elastic wave speed. The elastic wave speed
can be estimated based on known propagation distance and time. Linear regression was
applied for all successive peak axial displacements at each lateral position in the
corresponding spatial-temporal map to calculate the wave speed. Elastic wave mappings
across the whole eye were obtained using this method. In a previous study [120], a lateral
distance of 0.756 mm was used to reconstruct the elastic wave mappings of the porcine
67

ONH and PPS due to the high tissue elasticity. In this study, we adopted a lateral distance
to be 0.8 mm. Then the 3D data matrix was converted to 2D spatial-temporal maps at each
depth position, and these maps were interpolated in the time domain. Next, the time shift
was calculated using 1D cross-correlation algorithm in the lateral position with an interval
of 0.8 mm, and the elastic wave speed was calculated based on the known time shift and
lateral interval. This process was repeated in consecutive lateral positions and at each depth
position to reconstruct the elastic wave speed map. Finally, image segmentation was
performed, and unanticipated noise was removed. 3 x 3 medium and gaussian filters were
applied to increase the signal-to-noise ratio.

6.2.3 Phantom Preparation and System Synchronization
The gelatin phantom study was used to validate the accuracy of our UE system. Gelatin-
based (Gelatin G8-500, Fisher Scientific, USA) tissue-mimicking phantoms with equal
concentrations of silicon carbide powder (S5631, Sigma-Aldrich, St. Louis, MO, USA)
and sound scatters were fabricated. The stiffness of each phantom was controlled by the
gelatin concentration. The stiffness of the homogeneous phantom was first tested using our
UE system and then using the gold-standard – uniaxial mechanical testing (Model 5942,
Instron Corp., MA, USA). The mean and standard deviation were determined by measuring
each phantom five times using each method. Figure 6-3 (a) shows the schematic diagram
of the experimental set-up for the phantom study. The top and bottom boundaries of the
phantom were filled with a noncompressible medium, which in this case was water. The
68

rod of the shaker was placed perpendicular to the phantom to initiate the elastic wave
propagation and was aligned in the same plane with the array transducer to improve
accuracy.
Figure 6-3 (b) shows the synchronization diagram of the UE system. The UE system
was synchronized by an output trigger signal from the Vantage platform. The excitation
frequency of the shaker was set to 1 kHz with 1 cycle. The pulse repetition frequency of
10 kHz was used to enable the tracking of the elastic wave speed. 7 angles of the
compounding plane wave were used during the data acquisition. The shaker vibrated once
during each imaging experiment. The data acquisition time for the imaging array transducer
was 20 ms.

Figure 6-3. (a) Schematic diagram of the phantom study; (b) The synchronization sequence of ultrasonic
elastography system.

6.2.4 Animal Protocol
Surgery protocol was approved by the University of Southern California Institutional
Animal Care and Use Committee (IACUC). Prior to surgery, Dutch Belted Pigmented
rabbits (~ 2 kg) were anesthetized with ketamine (35mg/kg) and xylazine (5 mg/kg)
through subcutaneous injections. Additional anesthesia was induced using 2.5%
sevoflurane through a facial mask. 1 drop of 0.5% proparacaine hydrochloride ophthalmic
solution was administrated topically for ocular anesthesia. Heart rate, respiratory rate,
69

oxygen saturation, body temperature, non-invasive blood pressure, oxygen flow and end
tidal carbon dioxide (when sevoflurane was used) were recorded every 5 minutes while the
animals were unconscious.
Three rabbits were used in this study. Two 23 G trocars were used to create two ports
in the pars plana. An infusion line and a pressure sensor were inserted to control and read
the IOP levels respectively. The IOP level started at 10 mmHg and was raised to IOP level
of 30 mmHg in 5 mmHg increments.

6.3 Results
6.3.1 System Validation with Phantom Study
The phantom study was first conducted to verify the accuracy of our proposed UE
system. B-mode image was first acquired to determine the position of the phantom as
shown in the left of Figure 6-4 (a). Then elastography experiment was conducted, and the
spatial-temporal map was obtained after post processing as shown in the right of Figure 4
(a). The dashed line represents the wavefront of the elastic wave at each lateral position,
and the slope of this dashed line is the elastic wave speed. In this image, the elastic wave
speed is 3.38 m/s. Figure 6-4 (b) shows the elastic wave propagation in the phantom. The
dashed line represents the wavefront, with the elastic wave propagating from left to right.
After the elastography experiment, the stiffness of the same phantom was measured by
mechanical testing. The reconstructed Young’s modulus of the phantom by our UE system
is 34.34 ± 1.92 kPa, while the mechanical testing result is 38.70 ± 5.60 kPa. Our
elastographic result is comparable with the mechanical testing result, it verifies the
accuracy of our imaging system.

70

Figure 6-4. Phantom study results to validate the accuracy of the ultrasonic elastography system. (a) B-mode
imaging and the corresponding spatial-temporal map. The dashed line represents the wavefront of the elastic
wave; (b) The propagation of the elastic wave in the phantom with a time interval of 1 ms. The dashed line
represents the wavefront of the elastic wave, the wave propagates from left to right. Axial displacement unit:
µm

71

6.3.2 Results for Rabbit Eye

Figure 6-5. Spatial-temporal maps of the cornea, iris, lens, PPS and ONH at the IOP of 10 mmHg. Dashed
lines represent the wavefront of the elastic wave.

72

The spatial-temporal map, which represents the displacement curve in the lateral
direction as a function of the elastic wave propagation time, was first acquired to calculate
the elastic wave speed. Figure 6-5 shows a typical set of spatial-temporal maps for the
cornea, iris, lens, ONH and PPS at an IOP level of 10 mmHg. The slope of the wavefront,
which is represented by the black dotted line, is assumed to be the elastic wave speed,
which was calculated using the linear regression algorithm. Under this IOP level, the elastic
wave speed of cornea, iris, lens, ONH and PPS is 2.62 m/s, 1.01 m/s, 3.89 m/s, 1.63 m/s,
and 3.84 m/s respectively.

Figure 6-6. Statistical analysis of the Young’s modulus distributions of the ocular tissues with different IOP
levels.
Subsequently, the spatial-temporal maps were acquired and the elastic wave speed in
each tissue is measured under different IOP levels. The biomechanical property can be then
reconstructed using the Eq-2, where ⍴ is the density and c is the elastic wave speed. The
Young’s modulus is presented as mean ± standard deviation. As shown in Figure 6-6, at
73

IOP levels of 10, 15, 20, 25, and 30 mmHg, the Young’s modulus of cornea is 25.1 ± 0.9
kPa, 50.6 ± 7.2 kPa, 72.0 ± 8.6 kPa, 113.5 ± 9.1 kPa, and 177.2 ± 11.9 kPa respectively;
for the iris, it is 3.1 ± 0.7 kPa, 5.1 ± 1.0 kPa, 6.7 ± 0.9 kPa, 9.2 ± 1.2 kPa, and 11.9 ± 1.2
kPa respectively; for the lens, it is 54.8 ± 10.3 kPa, 58.9 ± 9.3 kPa, 54.2 ± 15.1 kPa, 61.9 ±
6.1 kPa, and 65.5 ± 7.0 kPa respectively; for the PPS, it is 43.6 ± 15.8 kPa, 105.0 ± 13.4
kPa, 184.7 ± 18.6 kPa, 375.9 ± 40.5 kPa, and 564.0 ± 50.7 kPa respectively; for the ONH,
it is 7.6 ± 2.2 kPa, 15.8 ± 3.3 kPa, 27.9 ± 2.1 kPa, 52.7 ± 3.2 kPa, 126.5 ± 5.3 kPa
respectively. The Young’s modulus measurement results in each tissue and IOP level were
evaluated using a Shapiro-Wilk test. The statistical analysis showed that Young’s modulus
measurements are distributed normally within each group of data. Differences in the
Young’s modulus between each tissue under different IOP levels were assessed using one-
way ANOVA. The statistical analysis showed the Young’s modulus of the cornea, iris, PPS
and ONH increased with increasing IOP levels with P ˂ 0.01.

Figure 6-7. The Young’s modulus distributions of the ocular tissues with different IOP levels and the
corresponding fittings. The dots represent the experimental results, the dashed lines represent the fittings
according to the experimental results.
74

. In Figure 6-7, The Young’s modulus of each tissue was plotted against IOP. The
Young’s modulus of the cornea and iris could be fitted with a linear curve. The Young’s
modulus of the PPS and ONH could be fitted with a second-order polynomial curve. The
Young’s modulus of the lens was not affected by changes in IOP.
Young’s modulus maps of the whole eye can be acquired by using cross-correlation
algorithm as shown in Figure 6-8. For better visualization, Figure 8 shows the B-mode
imaging and corresponding elastic wave speed distribution map of the whole eye at
different IOP levels. Although subtle morphologic changes could occur, they are difficult
to observe, while elastography can provide supplementary information regarding on the
biomechanical prospect.

Figure 6-8. B-mode imaging and the elastic wave velocity mapping of the same eye under different IOP
levels. The first column represents the B-mode imaging and the second column represents the corresponding
mapping.

75

6.4 Discussion and Conclusion
OCT and ultrasound have been the most applicable in the field of ophthalmology among
the range of imaging modalities due to their unique advantages. However, one limitation
is inability to image the whole eye at once. There are studies that used MRI to image the
whole eye. Ho et al. used the magic-angle enhancement effect to improve MRI sensitivity,
and Singh et al. designed novel radio-frequency coils to acquire images of the whole eye
[121, 122]. However, only structural information was provided with this method, such as
radius and curvature; there were no direct measurements to characterize the biomechanical
properties of anatomical structures. OCT has the highest imaging resolution among the
aforementioned imaging modalities and has been widely applied in both scientific research
and clinical care within ophthalmology. Due to the limited field of view, OCE has been
used to characterize the biomechanics of the single ocular tissue such as cornea [99], retina
[30, 123, 124], and ONH [33]. Recently, Li et at. developed a swept light source based
OCE to increase imaging depth, producing a system that can assess the biomechanics of
cornea and lens simultaneously [20]. Although the imaging penetration depth of OCE has
improved significantly, this system still cannot provide information on the biomechanical
properties of the whole eye.
Most ultrasonic elastography systems developed for ophthalmic applications utilize a
single-element transducer for imaging and capturing the elastic wave propagation [22, 27].
This has the advantage of high imaging resolution due to the high performance the
transducer related to high frequency and sensitivity. However, it has two primary
disadvantages: first, the imaging depth of the high frequency single-element transducer is
limited and cannot cover the entire range of the whole eye [120]; second, it cannot be easily
translated for clinic use since the mechanical scanning is slow, typically more than several
minutes per eye. The induced axial displacement is on the level of micrometers, and
movements by patients can significantly affect measurements, especially when scanning
occurs over such a long period. Therefore, using an ultrasonic array to serve as the imaging
transducer is an ideal choice. However, ultrasonic arrays usually suffer from the low
frequency due to the complexity of the fabrication process. Pekel et al. used a bandwidth
of 6-15 MHz array to evaluate the elasticity of ocular and periocular tissue after panretinal
photocoagulation [125]. This ultrasonic elastography system had the ability to image the
76

whole eye; however, imaging quality was not satisfactory due to the low frequency of the
array transducer. There have recently been many high frequency ultrasonic arrays available
to assemble the elastography system due to rapid advances in the manufacturing of array
transducers. Kwok et al. used a 50 MHz linear array transducer to reconstruct the
biomechanics of normal and keratoconus corneas [126]. Pavlatos et al. used a 55 MHz
transducer to reconstruct the strain map of the ONH [25]. These systems combine both
advantages of high imaging quality and fast scanning time, thereby providing accurate
biomechanics estimation. Utilizing high frequency arrays as the imaging transducers could
represent a significant advance in the field of ophthalmology. However, the imaging
penetration depth is still limiting its application to imaging the whole eye. In the ultrasound
elastography system proposed in this study, a moderate frequency range was selected to
enable both relatively high imaging quality and imaging penetration depth. This system
can accurately reconstruct the biomechanics of anatomical structures of the whole eye
simultaneously.
In this study, the biomechanical property of each ocular tissue is investigated at different
IOP levels simultaneously for the first time. The results show that our imaging system can
provide a meaningful tool to explore relationships among different ocular tissue at varying
levels of IOP. Our findings indicate that the biomechanics of the cornea and iris have a
linear relationship with IOP, the biomechanics of the PPS and ONH have a second-order
polynomial relationship with IOP, and the biomechanics of the lens are not affected by IOP.
Strip extensiometry revealed that the rabbit cornea exhibited a non-linear stress-strain
relationship when the load was over physiological conditions [127]. However, this
relationship could be linear if the load was within the physiological range [128].
Experimental ultrasonic micro-elastography and supersonic elastography also revealed the
linear relationship of Young’s modulus and IOP in a range of IOP that was similar to our
study [27, 129]. Similar to the cornea, the iris exhibited a non-linear stress-strain
relationship when the force load exceeded a certain range, while this relationship tended to
be linear when the force was below the physiologic threshold [130, 131]. Unlike with most
ocular tissues, evidence suggests that the biomechanics of lens is related more to the age
than IOP. Both UE and OCE revealed that the lens tends to be stiffer in aging animals [132,
133], while the influence of elevated IOP on lens biomechanics was found to be
77

insignificant [20, 132, 134]. Experimental results from tensile testing and elastography
revealed the non-linearity of ONH biomechanics in the physiological IOP range below 40
mmHg [33, 135]. The PPS also exhibited a similar non-linear relationship [120]. The
absolute values of measured Young’s modulus can be affected by different methods and
scales of the samples, boundary conditions, testing conditions, species of the animals, and
other factors. However, the results presented in this study are comparable with previously
reported findings and are consistent overall with the literature.
Shear wave based elastography is the dominant type of ultrasonic elastography in both
scientific research and clinical care. The group velocity is used to reconstruct tissue
biomechanics and is assumed to be correlated with the Young’s modulus under the
following assumptions: the imaging target is homogeneous, isotropic, large compared to
the wavelength of the induced shear wave, and purely elastic [136]. Using the group
velocity to reconstruct the Young’s modulus can cause errors when the imaging targets are
thin and viscous [137]. Most ocular tissues are thin and viscous; therefore, many models
have been developed to better estimate the biomechanics, such as Maxwell and Kevin-
Voigt model [27]. Shih et al. revealed that the Lamb wave model is suitable to reconstruct
the biomechanics of the cornea, since it exhibits thin layer, viscosity, and unique boundary
conditions [28]. However, there are no similar models for other ocular tissues. While
precise characterization of corneal biomechanics is important, the focus of this study is to
explore the interactions between each ocular tissue and IOP. In the future work, we expect
to have models available to estimate the biomechanics of the whole eye more accurately.
Adjust the height of the saline water column that is connected to the eye is a commonly
applied method to modulate IOP levels. This method can control IOP levels easily and
accurately, as demonstrated in this study. Qian et al. used this method to control IOP level,
and investigate changes in the biomechanical properties of the ex vivo ONH and PPS at
different IOP levels [120]. Their findings indicated that the biomechanical response
differed if IOP was increasing or decreasing, perhaps due to the change of the collagen
crimp structure [36, 120]. However, the IOP is difficult to lower from high levels once
achieved in vivo as the viscosity of the vitreous humor impedes the leakage of saline fluid
from the eye. Vitrectomy can be performed to replace the vitreous humor with saline fluid.
78

Therefore, future work could apply this approach to investigate how the biomechanical
response of the eye differs when IOP increases from low to high or decreases high to low.
In this study, a novel ultrasonic elastography system based on a moderate frequency
array and mechanical shaker was developed, and its application for ocular imaging was
demonstrated using rabbits as the animal model. The imaging system was first validated
by a gelatin imaging phantom. The Young’s modulus of the phantom was measured by our
imaging system and by mechanical testing; comparable results verified the accuracy of our
system. The frequency was carefully chosen to balance imaging resolution and penetration
depth. This system can cover the whole eye and provide information on the biomechanical
properties of multiple ocular tissues with high imaging quality. The IOP was manipulated
to simulate ocular hypertension and assess the relationships between biomechanical
properties and IOP. We conducted statistical analysis of each ocular tissue at different IOP
levels. B-mode imaging and corresponding biomechanical mapping at different IOP levels
was also conducted to visualize the effect of IOP changes on the whole eye. Each ocular
tissue exhibited a different response to the IOP changes; the cornea and iris had a linear
response, the PPS and ONH had a second-order polynomial response, and the lens had no
response. Owing to the short image acquisition time, our imaging system could be a useful
tool in research to investigate the biomechanics of the whole eye or in clinical care for
assessing risk of various ocular diseases, including keratoconus and glaucoma.

79

7. Chapter 7 Research the Anisotropy of the Equatorial Sclera
with Ultrasonic Elastography
7.1 Introduction
Sclera is a stiff fibrous tissue which expands from cornea to optic nerve head (ONH)
and covers the most neuropathy of the eye. It is the major force loading structure, prevent
the trauma from the external damage, and bear the dynamic internal intraocular pressure
fluctuation [138]. It leads to the rising attention to the biomechanical properties of the
sclera, as it plays a pivotal role to protect the eye. Currently, many researchers focused on
the biomechanical properties of the posterior sclera, for example, peripapillary sclera, due
to its importance for diagnosis the progression with the glaucoma [139]. The anisotropy of
the posterior sclera has been studied well to investigate mechanisms of the biomechanics
alteration during the progression of the glaucoma. However, the biomechanics and the
anisotropy of the equatorial sclera are also important since it serves as the protective layer
of the retina. Those properties are not only important to the glaucoma, but also many more
diseases in the ophthalmology.
Myopia is a prevalent slightness disorder in the worldwide affecting around one billion
people, especially in the developed area. The myopia is produced by the prolonging of the
vitreous chamber of the ocular globe axially. Mechanical testing study shows that the
alteration of the scleral biomechanics attributes to the shape change of the myopic eye
[140]. There remains a possible reason that the anisotropy of the equatorial sclera could
attribute for the elongating of the ocular globe in the axial direction instead of the azimuthal
direction. However, there is no known relevant study.
The anisotropy of the equatorial sclera could be also important to the early intervention
treatment of the ocular trauma. In the United States, it is estimated that 3% of the
emergency visit are for ocular trauma [141]. It requires early and rapid intervention to
prevent vision loss, especially for the open globe injuries as the rapid decreasing IOP could
lead to the severe retinal detachment [142]. In the clinic, the standard way to cure is to use
the suture, however, it is time- consuming. Researchers developed glue and adhesives for
easy-to-use purpose. The anisotropy property could attribute to improve the
biocompatibility of the proposed product.
80

To investigate the biomechanics of the sclera, various of methods have been used to
facilitate the study. Finite element modeling is a commonly used simulation method to
evaluate how the sclera behaves with different load [143]. It provides the additional tool
for numerical simulation with the experimental findings. There are many technologies
available to experimentally quantify the biomechanics of the sclera, such as
uniaxial/biaxial tensile testing [144], atomic force microscopy [14], polarized light
microscopy [145], magnetic resonance imaging [146] and elastography [120]. Among
which, elastography has the potential to reconstruct the biomechanics of the sclera
quantitively, with different IOP levels and intact eye structure.
Elastography is an emerging imaging modality which can quantify the biomechanics
quantitively and non-invasively. Optical coherence tomography (OCT) and ultrasonic
based elastography are the two common modalities in this field. Owing to the subtle
structures of the ocular tissue, OCE and high frequency ultrasonic elastography has been
widely applied in the ophthalmology. In this study, we aim to use the ultrasound
elastography to quantify the biomechanics of the equatorial sclera with 4 quadrants, which
are superior, inferior, nasal and temporal, and 2 directions which are equatorial direction
and anterior to posterior (AP) direction, and 5 different IOP levels starting from 10 mmHg
to 30 mmHg with 5 mmHg of increment. To achieve this goal, both biomechanical
mappings and statistical analysis are provided in this work, to the best of our knowledge,
it is the first time of the comprehensive study to characterize the biomechanical property
and isotropic property in the equatorial sclera.

7.2 Methods
The elastography imaging system set-up and IOP control system are the same as
described in the Chapter 6. The fresh pig eyes were acquired by the local slaughterhouse.
The conjunctiva and other affiliated tissues were removed by a blade. The porcine eyes
were then placed in a customized holder and maintained stable. The array and shaker were
placed in the same plane during each measurement. Figure 7-1 shows a typical scenario of
the measurement in the AP direction of the equatorial sclera. Figure shows the locations
of the superior, inferior, nasal and temporal quadrants. The elastography measurements
81

were conducted in the four quadrants in both AP and equatorial directions, IOP levels
ranging from 10 mmHg to 30 mmHg with 5 mmHg of increment, at least 3 measurements
were conducted.

Figure 7-1. (a) The diagram of the high frequency ultrasonic elastography configuration. (b) the locations
of the superior, inferior, nasal and temporal quadrants.

7.3 Results
The accuracy of the system was validated in the Chapter 6 with the phantom study. The
B-mode images were first acquired to determine the region of interest (ROI). The left
column of Figure 7-2 shows the B-mode imaging of the equatorial direction in the superior
quadrants under different IOP levels, the morphology change is barely discernible under
different IOP levels. Then the elastography measurements were conducted and the spatial-
temporal maps were acquired after post processing as shown as the right column of Figure
7-2. The color bar represents the arbitrary axial displacement. The red line represents the
wavefront of the elastic wave at each lateral position, the slope of this line represents the
elastic wave speed. The examples of the spatial-temporal maps are shown as the right
column of the Figure 7-2. The Young’s modulus can be reconstructed using Eq-2 and is
presented as mean ± standard derivation.
82

Figure 7-2. B-Mode Imaging and Spatial-Temporal Map in Different IOP at Superior along equatorial
direction.
83

Figure 7-3. B-Mode Imaging and Spatial-Temporal Map in Different IOP at Superior along AP direction.

84

Figure 7-4. Spatial-Temporal Map at two directions in Different IOP at Temporal quadrant.
85

Figure 7-5. Spatial-Temporal Map at two directions in Different IOP at Nasal quadrant.
86

Figure 7-6. Spatial-Temporal Map at two directions in Different IOP at Inferior quadrant.
87

Figure 7-7. The biomechanics mapping at the superior quadrant with two directions under different IOP
levels. For better visualization and enhance the image contrast, the elastic wave speed is used instead of the
Young’s modulus. The color bar represents the elastic wave speed, unit is m/s.
88

Figure 7-8. The biomechanics mapping at the temporal quadrant with two directions under different IOP
levels.
89

Figure 7-9. The biomechanics mapping at the nasal quadrant with two directions under different IOP levels.

90

Figure 7-10. The biomechanics mapping at the inferior quadrant with two directions under different IOP
levels.
91

From 10 mmHg to 30 mmHg with 5 mmHg of increment: In superior quadrant, the
Young’s modulus in the equatorial direction is 156.53 ± 41.69 kPa, 273.68 ± 82.61 kPa,
315.82 ± 25.72 kPa, 496.04± 21.92 kPa, 744.70 ± 67.90 kPa respectively, in the AP
direction is 8.96 ± 1.06 kPa, 15.04 ± 1.10 kPa, 24.23 ± 3.10 kPa, 45.52 ± 3.10 kPa, 71.04
± 2.06 kPa respectively; In temporal quadrant, the Young’s modulus in the equatorial
direction is 265.34 ± 17.05 kPa, 269.3 ± 19.19 kPa, 327.99 ± 44.45 kPa, 464.45 ± 16.42
kPa, 531.43 ± 63.02 kPa respectively, in the AP direction is 13.70 ± 0.4 kPa, 16.38 ± 0.33
kPa, 23.51 ± 0.51 kPa, 37.71 ± 0.39 kPa, 51.57 ± 2.19 kPa respectively; In temporal
quadrant, the Young’s modulus in the equatorial direction is 519.91 ± 14.70 kPa, 636.43 ±
110.12 kPa, 653.13 ± 105.13 kPa, 919.15 ± 116.16 kPa, 1113.03 ± 93.39 kPa respectively,
in the AP direction is 24.41 ± 2.54 kPa, 29.49 ± 2.67 kPa, 45.58 ± 5.45 kPa, 49.77 ± 2.91
kPa, 71.55 ± 2.6 kPa respectively; In inferior quadrant, the Young’s modulus in the
equatorial direction is 120.19 ± 9.45 kPa, 129.98 ± 13.13 kPa, 140.59 ± 11.70 kPa, 286.12
± 20.22 kPa, 330.11 ± 38.27 kPa respectively, in the AP direction is 39.05 ± 0.24 kPa,
72.65 ± 6.86 kPa, 106.37 ± 7.21 kPa, 158.84 ± 10.20 kPa, 278.22 ± 35.32 kPa respectively.
Figure 7-11 summarizes the above statistical analysis.

Figure 7-11. Statistical analysis of the Young’s modulus for all the measurements. The dashed line represents
the second polynomial fitting, error bar represents the standard derivation.
92

The value to characterize the anisotropy in each quadrant is defined by the ratio of the
mean Young’s modulus for each direction. From 10 mmHg to 30 mmHg with 5 mmHg of
increment, the value in the superior quadrant is 17.44, 18.20, 13.03, 10.90, 10.48
respectively; the value in the temporal quadrant is 19.37, 16.44, 13.95, 12.31, 10.30
respectively; the value in the nasal quadrant is 21.31, 21.58, 14.33, 18.47, 15.56
respectively; the value in the inferior quadrant is 3.08, 1.79, 1.32, 1.80, 1.19 respectively.

Figure 7-12. Statistical analysis of the anisotropy in different quadrants under each IOP level. The dashed
line represents the second polynomial fitting.

93

7.4 Discussion and Conclusion
Sclera serves as the major protective and force loading tissue, the biomechanics of the
sclera has been investigated through various methods. Due to the difficulty of directly
measuring the biomechanics of the sclera, finite element modeling is a common and
supplementary method to simulate how the biomechanical property of the sclera responds
to the elevated IOP. Although the conclusions could be helpful, some assumptions that this
method adopts are usually unrealistic, for example, assuming the rest of the sclera which
beyond the scope to be isotropic [143]. Uniaxial/biaxial mechanical testing have been used
to quantify the relation of the stress and the strain [13]. However, the sclera needs to be cut
into stripes, which lose the original boundary conditions which may affect the results.
Moreover, IOP can not be manipulated with the specimen strips. Inflation testing has the
ability to manipulate the IOP levels, however, it can only provide the relative strain change
[147].
Elastography is a non-invasive imaging modality which has the ability to reconstruct
the biomechanical properties quantitatively. The biomechanical properties of the equatorial
were evaluated in the four quadrants, two directions and under five different IOP levels.
There are mainly three findings: Firstly, our results indicate that the biomechanics of the
equatorial sclera has a second polynomial relationship with IOP. This result indicates that
the equatorial sclera tends to have a similar biomechanical response to the IOP with the
posterior sclera. Secondly, equatorial sclera is stiffer along with the equatorial directions,
and thus verifies the anisotropy. Finally, the anisotropy property is different in different
quadrants. The nasal quadrant equatorial sclera has the largest anisotropy property, while
the inferior quadrant has the lowest anisotropy. Moreover, the anisotropy property changes
with the different IOP levels, it tends to decrease while IOP level increases.
The anisotropy property has been verified in the posterior sclera [148]. The collagen
fibers alignment is considered to be the dominant factor which influences the anisotropy.
For example, there is the exitance of the ring shape collagen alignment around the ONH,
and the circumstantial direction is stiffer than the other direction. Evidence also shows that
the collagen fibers are different between posterior sclera and equatorial sclera [149]. The
alignment of the collagen fibers in the equatorial sclera is less directional, and the
uniformity of the collagen fibers is lower. The collagen fiber alignment could be helpful to
94

explain the anisotropy of the equatorial sclera. Further analysis such as polarized
microscopy and wide-angle X-ray scattering will be conducted to explain this phenomenon.
The anisotropy property differs as the position and IOP changes. Detailed micro-structural
analysis will be supplemented in this study to provide the explanation.
There are a few limitations in this study. In vitro porcine eyes are used as the study
object, and is lack of in vivo experiment. It is because the bulky ultrasonic array is difficult
to be mounted perpendicularly to the designated position. OCE has the potential to conduct
the in vivo experiment, and this method will be adopted in the future study. In each
quadrant, there are only two directions are measured instead of multi-angles. Since linear
array transducer is used as the imaging probe, changing the angle of the array is hard to be
manipulated. The 2D array could be a good solution, it can be controlled to image a certain
plane with anticipated directions.

95

8. Chapter 8 Summary and Future Work
8.1 Summary
In this study, high frequency single element ultrasonic transducer based UE, OCE and
array based UE system have been developed successfully to reconstruct the biomechanical
properties of the ocular tissues. Compared with ARF as the excitation method, using a
shaker as the vibration source is more favorable in ophthalmology, and its accuracy has
been proved in every elastography system with imaging phantoms. The high frequency UE
can achieve the resolution at micrometer level, and OCE can achieve sub-micrometer level,
these results show the potential of our technology to be applied into ophthalmology.
The biomechanical properties of ONH and PPS have been evaluated with high
frequency single element UE in time domain, both statistical analysis and point-to-point
Young’s modulus mapping have been provided. We further assessed the biomechanical
properties in ONH in frequency domain with OCE and an advanced Lamb wave model to
increase the accuracy. A shaker based OCE has been developed to investigate the
biomechanical properties of retina in living rabbits for the first time. The parameters of the
shaker have been optimized, and the accuracy has been tested. The implant surgeries of
retinal prosthetic electrode were performed and OCE system was used to evaluate how the
electrode affect the retinal biomechanics, supplement the data for a potential clinical trial.
The array-based system was applied to quantify the biomechanics of the major ocular
tissues in the whole eye range in vivo, and the point-to-point biomechanics mappings were
provided with different IOP levels. It has the potential to be translated into clinic as a
routine diagnostic method for glaucomatous patients. Finally, the anisotropy properties
were investigated using the array based elastography system. The biomechanics of superior,
inferior, nasal and temporal quadrants with AP direction and equatorial direction with
different IOP levels have been investigated. These findings have the potential to facilitate
the research in the glaucoma, myopia and ocular trauma.
In conclusion, we have demonstrated the development of the proposed elastography
system and their applications in ophthalmology at various objects and dimensions.

96

8.2 Future Work
The work presented in this proposal mainly demonstrate the feasibility of using UE and
OCE to reconstruct the biomechanical properties of ocular tissue in a favorable manner.
The future development of this work will focus on the following two areas:

8.2.1 2D Array-based Elastography System
In this proposal, a high frequency single element transducer based UE system has been
developed to investigate the biomechanical properties of ocular tissue. There are many
advantages of our high frequency UE system, for example, high frequency of the imaging
ultrasonic transducer enables high resolution images. However, there are drawbacks in our
high frequency UE system, one of them is the low imaging speed due to the mechanical
scanning. To improve the imaging speed, using array-based imaging system is a feasible
way. Array based elastography can emit the ultrasonic signal simultaneously, the
acquisition time could be reduced to hundreds of times. However, the array used in the
system is typically linear array, only 2D imaging can be provided. To accommodate the
need of 3D volumetric imaging, 2D ultrasonic array transducer is required. Figure 8-1
shows the programmable ultrasonic research platform and the commercial 2D array. We
have developed a high frequency 2D array with a center frequency of 12 MHz as shown in
Figure 8-2. this high frequency array transducer holds the potential to be applied in
elastography study. Future study will mainly focus on the advanced algorithm development,
including the data acquisition, data processing and imaging reconstruction.
97

Figure 8-1. Verasonics system and a 2D ultrasonic array transducer. (Source data from Verasonics website)

Figure 8-2. Homemade 12 MHz high frequency 2D array for elastography application.

8.2.2 Human Study
Previous study mainly focused on the application of the high resolution elastography
system in the cadaveric and animal study. We propose to evaluate how the progression of
98

the glaucoma affect the biomechanics of the eye in the whole globe range. The race, gender
and age will also be considered as the influencing factor. To achieve this goal, we have the
Institutional Review Boards (IRB) protocol approved. The screening stage was assembled
as shown in Figure 8-3. Flexibility is one of the important requirements as it can provide
the operator convenience. The stage has 10 degrees of freedom which allows the operator
to move the imaging probe at any dimension. Figure 8-2 demonstrates our preliminary
results: (a) shows the B-mode imaging of the eyelid, cornea, iris and lens, (b) is a typical
screening scenario, my colleague was screening my anterior eye, (c) the spatial-temporal
map of my iris, the elastic wave speed is around 1.01 m/s. The proposed system has the
advantage of easy to move as the whole system is assemble on a cart. Furthermore, the
proposed method does not need the anesthetics, no uncomfortable experience was found
during couple of the screenings.

Figure 8-3. The design of screening stage, 10 degrees of freedom.
99

Figure 8-4. (a) B-mode imaging of the anterior eye, (b) a typical screening scenario, (c) spatial-temporal
map of my iris.
100

9. Bibliography
[1] Y. S. Rabinowitz, "Keratoconus," (in English), Survey of Ophthalmology, vol. 42,
no. 4, pp. 297-319, Jan-Feb 1998.
[2] R. N. Weinreb and P. T. Khaw, "Primary open-angle glaucoma," (in English),
Lancet, vol. 363, no. 9422, pp. 1711-1720, May 22 2004.
[3] R. D. Jager, W. F. Mieler, and J. W. Miller, "Age-related macular degeneration -
Reply," (in English), New England Journal of Medicine, vol. 359, no. 16, pp. 1736-1736,
Oct 16 2008.
[4] L. Krishnan, J. B. Hoying, H. Nguyen, H. Song, and J. A. Weiss, "Interaction of
angiogenic microvessels with the extracellular matrix," (in English), American Journal of
Physiology-Heart and Circulatory Physiology, vol. 293, no. 6, pp. H3650-H3658, Dec
2007.
[5] B. R. Zhou, J. J. Chen, A. Kazemi, A. J. Sit, and X. M. Zhang, "An Ultrasound
Vibro-Elastography Technique for Assessing Papilledema," (in English), Ultrasound in
Medicine and Biology, vol. 45, no. 8, pp. 2034-2039, Aug 2019.
[6] H. A. Quigley and A. T. Broman, "The number of people with glaucoma worldwide
in 2010 and 2020," (in English), British Journal of Ophthalmology, vol. 90, no. 3, pp. 262-
267, Mar 2006.
[7] C. F. Burgoyne, J. C. Downs, A. J. Bellezza, J. K. F. Suh, and R. T. Hart, "The
optic nerve head as a biomechanical structure: a new paradigm for understanding the role
of IOP-related stress and strain in the pathophysiology of glaucomatous optic nerve head
damage," (in English), Progress in Retinal and Eye Research, vol. 24, no. 1, pp. 39-73, Jan
2005.
[8] I. A. Sigal, J. G. Flanagan, and C. R. Ethier, "Factors influencing optic nerve head
biomechanics," (in English), Investigative Ophthalmology & Visual Science, vol. 46, no.
11, pp. 4189-4199, Nov 2005.
101

[9] J. C. Morrison, E. C. Johnson, W. Cepurna, and L. J. Jia, "Understanding
mechanisms of pressure-induced optic nerve damage," (in English), Progress in Retinal
and Eye Research, vol. 24, no. 2, pp. 217-240, Mar 2005.
[10] H. A. Quigley, "The contribution of the sclera and lamina cribrosa to the
pathogenesis of glaucoma: diagnostic and treatment implications," (in English), New
Trends in Basic and Clinical Research of Glaucoma: A Neurodegenerative Disease of the
Visual System, Pt A, vol. 220, pp. 59-86, 2015.
[11] I. C. Campbell, B. Coudrillier, and C. R. Ethier, "Biomechanics of the Posterior
Eye: A Critical Role in Health and Disease," (in English), Journal of Biomechanical
Engineering-Transactions of the Asme, vol. 136, no. 2, Feb 2014.
[12] E. Spoerl, A. G. Boehm, and L. E. Pillunat, "The influence of various substances
on the biomechanical behavior of lamina cribrosa and peripapillary sclera," (in English),
Investigative Ophthalmology & Visual Science, vol. 46, no. 4, pp. 1286-1290, Apr 2005.
[13] A. Eilaghi, J. G. Flanagan, I. Tertinegg, C. A. Simmons, G. W. Brodland, and C. R.
Ethier, "Biaxial mechanical testing of human sclera," (in English), Journal of
Biomechanics, vol. 43, no. 9, pp. 1696-1701, Jun 18 2010.
[14] C. Braunsmann, C. M. Hammer, J. Rheinlaender, F. E. Kruse, T. E. Schaffer, and
U. Schlotzer-Schrehardt, "Evaluation of Lamina Cribrosa and Peripapillary Sclera
Stiffness in Pseudoexfoliation and Normal Eyes by Atomic Force Microscopy," (in
English), Investigative Ophthalmology & Visual Science, vol. 53, no. 6, pp. 2960-2967,
May 2012.
[15] B. Coudrillier, J. Tian, S. Alexander, K. M. Myers, H. A. Quigley, and T. D.
Nguyen, "Biomechanics of the Human Posterior Sclera: Age- and Glaucoma-Related
Changes Measured Using Inflation Testing," (in English), Investigative Ophthalmology &
Visual Science, vol. 53, no. 4, pp. 1714-1728, Apr 2012.
[16] R. Muthupillai, D. J. Lomas, P. J. Rossman, J. F. Greenleaf, A. Manduca, and R. L.
Ehman, "Magnetic-Resonance Elastography by Direct Visualization of Propagating
102

Acoustic Strain Waves," (in English), Science, vol. 269, no. 5232, pp. 1854-1857, Sep 29
1995.
[17] J. Ophir, I. Cespedes, H. Ponnekanti, Y. Yazdi, and X. Li, "Elastography - a
Quantitative Method for Imaging the Elasticity of Biological Tissues," (in English),
Ultrasonic Imaging, vol. 13, no. 2, pp. 111-134, Apr 1991.
[18] D. Huang et al., "Optical Coherence Tomography," (in English), Science, vol. 254,
no. 5035, pp. 1178-1181, Nov 22 1991.
[19] Y. Qu et al., "Acoustic Radiation Force Optical Coherence Elastography of Corneal
Tissue," IEEE J Sel Top Quantum Electron, vol. 22, no. 3, May-Jun 2016.
[20] Y. Li et al., "Simultaneously imaging and quantifying in vivo mechanical properties
of crystalline lens and cornea using optical coherence elastography with acoustic radiation
force excitation," (in English), Apl Photonics, vol. 4, no. 10, Oct 2019.
[21] Y. M. He et al., "Confocal Shear Wave Acoustic Radiation Force Optical
Coherence Elastography for Imaging and Quantification of the In Vivo Posterior Eye," (in
English), Ieee Journal of Selected Topics in Quantum Electronics, vol. 25, no. 1, Jan-Feb
2019.
[22] X. J. Qian, T. Ma, M. Y. Yu, X. Y. Chen, K. K. Shung, and Q. F. Zhou, "Multi-
functional Ultrasonic Micro-elastography Imaging System," (in English), Scientific
Reports, vol. 7, Apr 27 2017.
[23] X. Qian et al., "Ultrasonic Microelastography to Assess Biomechanical Properties
of the Cornea," IEEE Trans Biomed Eng, vol. 66, no. 3, pp. 647-655, Mar 2019.
[24] S. K. Alam, D. W. Richards, and K. J. Parker, "Detection of Intraocular-Pressure
Change in the Eye Using Sonoelastic Doppler Ultrasound," (in English), Ultrasound in
Medicine and Biology, vol. 20, no. 8, pp. 751-758, 1994.
[25] E. Pavlatos, Y. H. Ma, K. Clayson, X. L. Pan, and J. Liu, "Regional Deformation
of the Optic Nerve Head and Peripapillary Sclera During IOP Elevation," (in English),
Investigative Ophthalmology & Visual Science, vol. 59, no. 8, pp. 3779-3788, Jul 2018.
103

[26] Y. H. Ma et al., "Mechanical Deformation of Human Optic Nerve Head and
Peripapillary Tissue in Response to Acute IOP Elevation," (in English), Investigative
Ophthalmology & Visual Science, vol. 60, no. 4, pp. 913-920, Mar 2019.
[27] X. J. Qian et al., "Ultrasonic Microelastography to Assess Biomechanical
Properties of the Cornea," (in English), Ieee Transactions on Biomedical Engineering, vol.
66, no. 3, pp. 647-655, Mar 2019.
[28] C. C. Shih et al., "Quantitative Assessment of Thin-Layer Tissue Viscoelastic
Properties Using Ultrasonic Micro-Elastography With Lamb Wave Model," (in English),
Ieee Transactions on Medical Imaging, vol. 37, no. 8, pp. 1887-1898, Aug 2018.
[29] M. A. Kirby et al., "Optical coherence elastography in ophthalmology," (in
English), Journal of Biomedical Optics, vol. 22, no. 12, Dec 2017.
[30] X. J. Qian et al., "In vivo evaluation of posterior eye elasticity using shaker-based
optical coherence elastography," (in English), Experimental Biology and Medicine, vol.
245, no. 4, pp. 282-288, Feb 2020.
[31] G. F. Pinton, J. J. Dahl, and G. E. Trahey, "Rapid tracking of small displacements
with ultrasound," (in English), Ieee Transactions on Ultrasonics Ferroelectrics and
Frequency Control, vol. 53, no. 6, pp. 1103-1117, Jun 2006.
[32] M. L. Palmeri, M. H. Wang, J. J. Dahl, K. D. Frinkley, and K. R. Nightingale,
"Quantifying hepatic shear modulus in vivo using acoustic radiation force," (in English),
Ultrasound in Medicine and Biology, vol. 34, no. 4, pp. 546-558, Apr 2008.
[33] Z. D. Du et al., "Quantitative confocal optical coherence elastography for
evaluating biomechanics of optic nerve head using Lamb wave model," (in English),
Neurophotonics, vol. 6, no. 4, Oct-Dec 2019.
[34] N. J. Jan, K. Lathrop, and I. A. Sigal, "Collagen Architecture of the Posterior Pole:
High-Resolution Wide Field of View Visualization and Analysis Using Polarized Light
Microscopy," Invest Ophthalmol Vis Sci, vol. 58, no. 2, pp. 735-744, Feb 1 2017.
104

[35] N. J. Jan et al., "Microstructural Crimp of the Lamina Cribrosa and Peripapillary
Sclera Collagen Fibers," Invest Ophthalmol Vis Sci, vol. 58, no. 9, pp. 3378-3388, Jul 1
2017.
[36] N. J. Jan and I. A. Sigal, "Collagen fiber recruitment: A microstructural basis for
the nonlinear response of the posterior pole of the eye to increases in intraocular pressure,"
Acta Biomater, vol. 72, pp. 295-305, May 2018.
[37] M. J. A. Girard, J. K. F. Suh, M. Bottlang, C. F. Burgoyne, and J. C. Downs,
"Scleral Biomechanics in the Aging Monkey Eye," (in English), Investigative
Ophthalmology & Visual Science, vol. 50, no. 11, pp. 5226-5237, Nov 2009.
[38] M. J. A. Girard, J. C. Downs, C. F. Burgoyne, and J. K. F. Suh, "Experimental
surface strain mapping of porcine peripapillary sclera due to elevations of intraocular
pressure," (in English), Journal of Biomechanical Engineering-Transactions of the Asme,
vol. 130, no. 4, Aug 2008.
[39] J. C. Downs, M. D. Roberts, and C. F. Burgoyne, "Mechanical environment of the
optic nerve head in glaucoma," (in English), Optometry and Vision Science, vol. 85, no. 6,
pp. 425-435, Jun 2008.
[40] E. M. Vanbuskirk and G. A. Cioffi, "Glaucomatous Optic Neuropathy," (in
English), American Journal of Ophthalmology, vol. 113, no. 4, pp. 447-452, Apr 15 1992.
[41] W. Wang, M. He, Z. H. Li, and W. Y. Huang, "Epidemiological variations and
trends in health burden of glaucoma worldwide," (in English), Acta Ophthalmologica, vol.
97, no. 3, pp. E349-E355, May 2019.
[42] C. Burgoyne, "The morphological difference between glaucoma and other optic
neuropathies," J Neuroophthalmol, vol. 35 Suppl 1, pp. S8-S21, Sep 2015.
[43] M. Hidalgo-Aguirre, J. Gitelman, M. R. Lesk, and S. Costantino, "Automatic
segmentation of the optic nerve head for deformation measurements in video rate optical
coherence tomography," (in English), Journal of Biomedical Optics, vol. 20, no. 11, Nov
2015.
105

[44] A. J. Bellezza, R. T. Hart, and C. F. Burgoyne, "The optic nerve head as a
biomechanical structure: Initial finite element modeling," (in English), Investigative
Ophthalmology & Visual Science, vol. 41, no. 10, pp. 2991-3000, Sep 2000.
[45] J. Albon, P. P. Purslow, W. S. S. Karwatowski, and D. L. Easty, "Age related
compliance of the lamina cribrosa in human eyes," (in English), British Journal of
Ophthalmology, vol. 84, no. 3, pp. 318-323, Mar 2000.
[46] K. S. Yadav, R. Rajpurohit, and S. Sharma, "Glaucoma: Current treatment and
impact of advanced drug delivery systems," (in English), Life Sciences, vol. 221, pp. 362-
376, Mar 15 2019.
[47] Y. Hua, A. P. Voorhees, and I. A. Sigal, "Cerebrospinal Fluid Pressure: Revisiting
Factors Influencing Optic Nerve Head Biomechanics," Invest Ophthalmol Vis Sci, vol. 59,
no. 1, pp. 154-165, Jan 1 2018.
[48] Z. H. Wu, G. H. Xu, R. N. Weinreb, M. Yu, and C. K. S. Leung, "Optic Nerve Head
Deformation in Glaucoma A Prospective Analysis of Optic Nerve Head Surface and
Lamina Cribrosa Surface Displacement," (in English), Ophthalmology, vol. 122, no. 7, pp.
1317-1329, Jul 2015.
[49] E. A. Sander, J. C. Downs, R. T. Hart, C. F. Burgoyne, and E. A. Nauman, "A
cellular solid model of the lamina cribrosa: Mechanical dependence on morphology," (in
English), Journal of Biomechanical Engineering-Transactions of the Asme, vol. 128, no.
6, pp. 879-889, Dec 2006.
[50] J. D. Pyne, K. Genovese, L. Casaletto, and J. P. Vande Geest, "Sequential-Digital
Image Correlation for Mapping Human Posterior Sclera and Optic Nerve Head
Deformation," (in English), Journal of Biomechanical Engineering-Transactions of the
Asme, vol. 136, no. 2, Feb 2014.
[51] Z. L. Han et al., "Optical coherence elastography assessment of corneal
viscoelasticity with a modified Rayleigh-Lamb wave model," (in English), Journal of the
Mechanical Behavior of Biomedical Materials, vol. 66, pp. 87-94, Feb 2017.
106

[52] V. Y. Zaitsev et al., "Optical coherence elastography for strain dynamics
measurements in laser correction of cornea shape," (in English), Journal of Biophotonics,
vol. 10, no. 11, pp. 1450-1463, Nov 2017.
[53] J. S. Li, Z. L. Han, M. Singh, M. D. Twa, and K. V. Larin, "Differentiating untreated
and cross-linked porcine corneas of the same measured stiffness with optical coherence
elastography," (in English), Journal of Biomedical Optics, vol. 19, no. 11, Nov 2014.
[54] Z. L. Han et al., "Quantitative methods for reconstructing tissue biomechanical
properties in optical coherence elastography: a comparison study," (in English), Physics in
Medicine and Biology, vol. 60, no. 9, pp. 3531-3547, May 7 2015.
[55] L. Ambrozinski et al., "Acoustic micro-tapping for non-contact 4D imaging of
tissue elasticity," (in English), Scientific Reports, vol. 6, Dec 23 2016.
[56] S. G. Chen, M. Fatemi, and J. F. Greenleaf, "Quantifying elasticity and viscosity
from measurement of shear wave speed dispersion," (in English), Journal of the Acoustical
Society of America, vol. 115, no. 6, pp. 2781-2785, Jun 2004.
[57] Y. Q. Qu et al., "In Vivo Elasticity Mapping of Posterior Ocular Layers Using
Acoustic Radiation Force Optical Coherence Elastography," (in English), Investigative
Ophthalmology & Visual Science, vol. 59, no. 1, pp. 455-461, Jan 2018.
[58] I. Z. Nenadic, M. W. Urban, M. Bernal, and J. F. Greenleaf, "Phase velocities and
attenuations of shear, Lamb, and Rayleigh waves in plate-like tissues submerged in a fluid
(L)," J Acoust Soc Am, vol. 130, no. 6, pp. 3549-52, Dec 2011.
[59] T. M. Nguyen, M. Couade, J. Bercoff, and M. Tanter, "Assessment of Viscous and
Elastic Properties of Sub-Wavelength Layered Soft Tissues Using Shear Wave
Spectroscopy: Theoretical Framework and In Vitro Experimental Validation," (in English),
Ieee Transactions on Ultrasonics Ferroelectrics and Frequency Control, vol. 58, no. 11,
pp. 2305-2315, Nov 2011.
[60] J. Zhu et al., "Imaging and characterizing shear wave and shear modulus under
orthogonal acoustic radiation force excitation using OCT Doppler variance method," (in
English), Optics Letters, vol. 40, no. 9, pp. 2099-2102, May 1 2015.
107

[61] J. Zhang, B. Rao, L. F. Yu, and Z. P. Chen, "High-dynamic-range quantitative
phase imaging with spectral domain phase microscopy," (in English), Optics Letters, vol.
34, no. 21, pp. 3442-3444, Nov 1 2009.
[62] A. Loewenstein, "The significance of early detection of age-related macular
degeneration," (in English), Retina-the Journal of Retinal and Vitreous Diseases, vol. 27,
no. 7, pp. 873-878, Sep 2007.
[63] H. R. Novotny and D. L. Alvis, "A method of photographing fluorescence in
circulating blood in the human retina," Circulation, vol. 24, pp. 82-6, Jul 1961.
[64] P. P. Connell et al., "Risk factors for age-related maculopathy," J Ophthalmol, vol.
2009, p. 360764, 2009.
[65] G. Age-Related Eye Disease Study Research, "The Age-Related Eye Disease Study
system for classifying age-related macular degeneration from stereoscopic color fundus
photographs: the Age-Related Eye Disease Study Report Number 6," Am J Ophthalmol,
vol. 132, no. 5, pp. 668-81, Nov 2001.
[66] M. R. Hee et al., "Optical coherence tomography of age-related macular
degeneration and choroidal neovascularization," Ophthalmology, vol. 103, no. 8, pp. 1260-
70, Aug 1996.
[67] L. Krishnan, J. B. Hoying, H. Nguyen, H. Song, and J. A. Weiss, "Interaction of
angiogenic microvessels with the extracellular matrix," Am J Physiol Heart Circ Physiol,
vol. 293, no. 6, pp. H3650-8, Dec 2007.
[68] E. T. Detorakis, E. E. Drakonaki, M. K. Tsilimbaris, I. G. Pallikaris, and S.
Giarmenitis, "Real-time ultrasound elastographic imaging of ocular and periocular tissues:
a feasibility study," Ophthalmic Surg Lasers Imaging, vol. 41, no. 1, pp. 135-41, Jan-Feb
2010.
[69] D. V. Litwiller et al., "MR Elastography of the Ex Vivo Bovine Globe," (in English),
Journal of Magnetic Resonance Imaging, vol. 32, no. 1, pp. 44-51, Jul 2010.
108

[70] K. V. Larin and D. D. Sampson, "Optical coherence elastography - OCT at work in
tissue biomechanics [Invited]," Biomed Opt Express, vol. 8, no. 2, pp. 1172-1202, Feb 1
2017.
[71] J. S. Li et al., "Air-pulse OCE for assessment of age-related changes in mouse
cornea in vivo," (in English), Laser Physics Letters, vol. 11, no. 6, Jun 1 2014.
[72] S. Wang et al., "Noncontact measurement of elasticity for the detection of soft-
tissue tumors using phase-sensitive optical coherence tomography combined with a
focused air-puff system," (in English), Optics Letters, vol. 37, no. 24, pp. 5184-5186, Dec
15 2012.
[73] W. J. Qi et al., "Phase-resolved acoustic radiation force optical coherence
elastography," (in English), Journal of Biomedical Optics, vol. 17, no. 11, Nov 2012.
[74] T. M. Nguyen, B. Arnal, S. Z. Song, Z. H. Huang, R. K. Wang, and M. O'Donnell,
"Shear wave elastography using amplitude-modulated acoustic radiation force and phase-
sensitive optical coherence tomography," (in English), Journal of Biomedical Optics, vol.
20, no. 1, Jan 2015.
[75] E. Mikula, K. Hollman, D. Chai, J. V. Jester, and T. Juhasz, "Measurement of
Corneal Elasticity with an Acoustic Radiation Force Elasticity Microscope," (in English),
Ultrasound in Medicine and Biology, vol. 40, no. 7, pp. 1671-1679, Jul 2014.
[76] A. J. Sit, S. C. Lin, A. Kazemi, J. W. McLaren, C. M. Pruet, and X. M. Zhang, "In
Vivo Noninvasive Measurement of Young's Modulus of Elasticity in Human Eyes: A
Feasibility Study," (in English), Journal of Glaucoma, vol. 26, no. 11, pp. 967-973, Nov
2017.
[77] M. L. Palmeri, Y. F. Deng, N. C. Rouze, and K. R. Nightingale, "Dependence of
shear wave spectral content on acoustic radiation force excitation duration and spatial
beamwidth," (in English), 2014 Ieee International Ultrasonics Symposium (Ius), pp. 1105-
1108, 2014.
[78] E. Widman, E. Maksuti, C. Amador, M. W. Urban, K. Caidahl, and M. Larsson,
"Shear Wave Elastography Quantifies Stiffness in Ex Vivo Porcine Artery with Stiffened
109

Arterial Region," (in English), Ultrasound in Medicine and Biology, vol. 42, no. 10, pp.
2423-2435, Oct 2016.
[79] K. Chen, A. P. Rowley, and J. D. Weiland, "Elastic properties of porcine ocular
posterior soft tissues," (in English), Journal of Biomedical Materials Research Part A, vol.
93a, no. 2, pp. 634-645, May 2010.
[80] M. S. Humayun, E. deJuan, G. Dagnelie, R. J. Greenberg, R. H. Prost, and D. H.
Phillips, "Visual perception elicited by electrical stimulation of retina in blind humans," (in
English), Archives of Ophthalmology, vol. 114, no. 1, pp. 40-46, Jan 1996.
[81] D. T. Hartong, E. L. Berson, and T. P. Dryja, "Retinitis pigmentosa," (in English),
Lancet, vol. 368, no. 9549, pp. 1795-1809, Nov 18 2006.
[82] R. D. Jager, W. F. Mieler, and J. W. Miller, "Medical progress: Age-related macular
degeneration," (in English), New England Journal of Medicine, vol. 358, no. 24, pp. 2606-
2617, Jun 12 2008.
[83] M. S. Humayun et al., "Visual perception in a blind subject with a chronic
microelectronic retinal prosthesis," (in English), Vision Research, vol. 43, no. 24, pp. 2573-
2581, Nov 2003.
[84] M. Humayun, R. Propst, E. Dejuan, K. Mccormick, and D. Hickingbotham,
"Bipolar Surface Electrical-Stimulation of the Vertebrate Retina," (in English), Archives of
Ophthalmology, vol. 112, no. 1, pp. 110-116, Jan 1994.
[85] M. S. Humayun et al., "Pattern electrical stimulation of the human retina," (in
English), Vision Research, vol. 39, no. 15, pp. 2569-2576, Jul 1999.
[86] J. D. Weiland, W. T. Liu, and M. S. Humayun, "Retinal prosthesis," (in English),
Annual Review of Biomedical Engineering, vol. 7, pp. 361-401, 2005.
[87] Y. H. L. Luo and L. da Cruz, "The Argus (R) II Retinal Prosthesis System," (in
English), Progress in Retinal and Eye Research, vol. 50, pp. 89-107, Jan 2016.
110

[88] M. S. Humayun et al., "Interim Results from the International Trial of Second
Sight's Visual Prosthesis," (in English), Ophthalmology, vol. 119, no. 4, pp. 779-788, Apr
2012.
[89] L. da Cruz et al., "The Argus II epiretinal prosthesis system allows letter and word
reading and long-term function in patients with profound vision loss," (in English), British
Journal of Ophthalmology, vol. 97, no. 5, pp. 632-636, May 2013.
[90] D. D. Zhou, J. D. Dorn, R. J. Greenberg, and A. I. S. Grp, "The Argus (R) Ii Retinal
Prosthesis System: An Overview," (in English), Electronic Proceedings of the 2013 Ieee
International Conference on Multimedia and Expo Workshops (Icmew), 2013.
[91] D. Yanai, J. D. Weiland, M. Mahadevappa, R. J. Greenberg, I. Fine, and M. S.
Humayun, "Visual performance using a retinal prosthesis in three subjects with retinitis
pigmentosa," (in English), American Journal of Ophthalmology, vol. 143, no. 5, pp. 820-
827, May 2007.
[92] M. Mahadevappa, J. D. Weiland, R. J. Greenberg, and M. S. Humayun, "Perceptual
thresholds and electrode impedance in three retinal prosthesis subjects," (in English), Ieee
Transactions on Neural Systems and Rehabilitation Engineering, vol. 13, no. 2, pp. 201-
206, Jun 2005.
[93] M. S. Humayun et al., "Towards a completely implantable, light-sensitive
intraocular retinal prosthesis.," (in English), Proceedings of the 23rd Annual International
Conference of the Ieee Engineering in Medicine and Biology Society, Vols 1-4, vol. 23, pp.
3422-3425, 2001.
[94] L. da Cruz et al., "Five-Year Safety and Performance Results from the Argus II
Retinal Prosthesis System Clinical Trial," (in English), Ophthalmology, vol. 123, no. 10,
pp. 2248-2254, Oct 2016.
[95] L. Yue et al., "Ten-Year Follow-up of a Blind Patient Chronically Implanted with
Epiretinal Prosthesis Argus I," (in English), Ophthalmology, vol. 122, no. 12, pp. 2545-+,
Dec 2015.
111

[96] H. C. Stronks and G. Dagnelie, "The functional performance of the Argus II retinal
prosthesis," Expert Rev Med Devices, vol. 11, no. 1, pp. 23-30, Jan 2014.
[97] G. Dagnelie et al., "Performance of real-world functional vision tasks by blind
subjects improves after implantation with the Argus(R) II retinal prosthesis system," Clin
Exp Ophthalmol, vol. 45, no. 2, pp. 152-159, Mar 2017.
[98] E. Pavlatos, X. L. Pan, K. Clayson, R. T. Hart, P. Weber, and J. Liu, "ONH
Deformation in Human Eyes Using Ultrasound Speckle Tracking," (in English),
Investigative Ophthalmology & Visual Science, vol. 58, no. 8, Jun 2017.
[99] Z. Jin et al., "In-vivo 3D corneal elasticity using air-coupled ultrasound optical
coherence elastography," (in English), Biomedical Optics Express, vol. 10, no. 12, pp.
6272-6285, Dec 1 2019.
[100] Y. Li, S. Moon, J. J. Chen, Z. K. Zhu, and Z. P. Chen, "Ultrahigh-sensitive optical
coherence elastography," (in English), Light-Science & Applications, vol. 9, no. 1, Apr 13
2020.
[101] B. Zhou, J. J. Chen, A. Kazemi, A. J. Sit, and X. Zhang, "An Ultrasound Vibro-
Elastography Technique for Assessing Papilledema," Ultrasound Med Biol, vol. 45, no. 8,
pp. 2034-2039, Aug 2019.
[102] D. C. Rodger, J. D. Weiland, M. S. Humayun, and Y. C. Tai, "Scalable high lead-
count parylene package for retinal prostheses," (in English), Sensors and Actuators B-
Chemical, vol. 117, no. 1, pp. 107-114, Sep 12 2006.
[103] Q. Zeng, S. S. Zhao, H. G. Yang, Y. Zhang, and T. Z. Wu, "Micro/Nano
Technologies for High-Density Retinal Implant," (in English), Micromachines, vol. 10, no.
6, Jun 2019.
[104] K. Xia, B. Sun, Q. Zeng, T. Z. Wu, and M. S. Humayun, "Surface Modification of
Neural Stimulating/Recording Microelectrodes with High-Performance Platinum-Pillar
Coatings," (in English), 2017 Ieee 12th International Conference on Nano/Micro
Engineered and Molecular Systems (Nems), pp. 291-294, 2017.
112

[105] S. S. Park et al., "Posterior Segment Complications after Vitrectomy for Macular
Hole," (in English), Ophthalmology, vol. 102, no. 5, pp. 775-781, May 1995.
[106] O. P. Gupta et al., "Postoperative complications associated with 25-gauge pars
plana vitrectomy," (in English), Ophthalmic Surgery Lasers & Imaging, vol. 38, no. 4, pp.
270-275, Jul-Aug 2007.
[107] Z. L. Han et al., "Quantitative assessment of corneal viscoelasticity using optical
coherence elastography and a modified Rayleigh-Lamb equation," (in English), Journal of
Biomedical Optics, vol. 20, no. 2, Feb 2015.
[108] F. Zvietcovich, P. Pongchalee, P. Meemon, J. P. Rolland, and K. J. Parker,
"Reverberant 3D optical coherence elastography maps the elasticity of individual corneal
layers," (in English), Nature Communications, vol. 10, Oct 25 2019.
[109] O. Bergamin, A. Schoetzau, K. Sugimoto, and M. Zulauf, "The influence of iris
color on the pupillary light reflex," (in English), Graefes Archive for Clinical and
Experimental Ophthalmology, vol. 236, no. 8, pp. 567-570, Aug 1998.
[110] R. H. Masland, "The fundamental plan of the retina," (in English), Nature
Neuroscience, vol. 4, no. 9, pp. 877-886, Sep 2001.
[111] I. A. Sigal and C. R. Ethier, "Biomechanics of the optic nerve head," (in English),
Experimental Eye Research, vol. 88, no. 4, pp. 799-807, Apr 2009.
[112] J. K. Pijanka et al., "Depth-Dependent Changes in Collagen Organization in the
Human Peripapillary Sclera," (in English), Plos One, vol. 10, no. 2, Feb 25 2015.
[113] G. S. Ang, F. Bochmann, J. Townend, and A. Azuara-Blanco, "Corneal
biomechanical properties in primary open angle glaucoma and normal tension glaucoma,"
(in English), Journal of Glaucoma, vol. 17, no. 4, pp. 259-262, Jun-Jul 2008.
[114] N. A. McBrien, A. I. Jobling, and A. Gentle, "Biomechanics of the Sclera in Myopia:
Extracellular and Cellular Factors," (in English), Optometry and Vision Science, vol. 86,
no. 1, pp. 23-30, Jan 2009.
113

[115] A. Saad, Y. Lteif, E. Azan, and D. Gatinel, "Biomechanical Properties of
Keratoconus Suspect Eyes," (in English), Investigative Ophthalmology & Visual Science,
vol. 51, no. 6, pp. 2912-2916, Jun 2010.
[116] M. Singh, A. Nair, S. R. Aglyamov, and K. V. Larin, "Compressional Optical
Coherence Elastography of the Cornea," (in English), Photonics, vol. 8, no. 4, Apr 2021.
[117] S. Kwok et al., "Heartbeat-Induced Corneal Axial Displacement and Strain
Measured by High Frequency Ultrasound Elastography in Human Volunteers," Transl Vis
Sci Technol, vol. 9, no. 13, p. 33, Dec 2020.
[118] C. C. Weng, P. Y. Chen, D. Chou, C. C. Shih, and C. C. Huang, "High Frequency
Ultrasound Elastography for Estimating the Viscoelastic Properties of the Cornea Using
Lamb Wave Model," (in English), Ieee Transactions on Biomedical Engineering, vol. 68,
no. 9, pp. 2637-2644, Sep 2021.
[119] Y. F. Deng, N. C. Rouze, M. L. Palmeri, and K. R. Nightingale, "Ultrasonic Shear
Wave Elasticity Imaging Sequencing and Data Processing Using a Verasonics Research
Scanner," (in English), Ieee Transactions on Ultrasonics Ferroelectrics and Frequency
Control, vol. 64, no. 1, pp. 164-176, Jan 2017.
[120] X. J. Qian et al., "Ultrasonic elastography to assess biomechanical properties of the
optic nerve head and peripapillary sclera of the eye," (in English), Ultrasonics, vol. 110,
Feb 2021.
[121] L. C. Ho et al., "Magic angle-enhanced MRI of fibrous microstructures in sclera
and cornea with and without intraocular pressure loading," Invest Ophthalmol Vis Sci, vol.
55, no. 9, pp. 5662-72, Aug 7 2014.
[122] K. D. Singh, N. S. Logan, and B. Gilmartin, "Three-dimensional modeling of the
human eye based on magnetic resonance imaging," (in English), Investigative
Ophthalmology & Visual Science, vol. 47, no. 6, pp. 2272-2279, Jun 2006.
[123] Y. Q. Qu et al., "Quantified elasticity mapping of retinal tissue using acoustic
radiation force optical coherence elastography," (in English), Investigative Ophthalmology
& Visual Science, vol. 58, no. 8, Jun 2017.
114

[124] R. Z. Li et al., "High resolution optical coherence elastography of retina under
prosthetic electrode," (in English), Quantitative Imaging in Medicine and Surgery, vol. 11,
no. 3, pp. 918-927, Mar 2021.
[125] G. Pekel, K. Agladioglu, S. Acer, R. Yagci, and A. Kasikci, "Evaluation of Ocular
and Periocular Elasticity after Panretinal Photocoagulation: An Ultrasonic Elastography
Study," (in English), Current Eye Research, vol. 40, no. 3, pp. 332-337, Mar 2015.
[126] S. Kwok, N. Hazen, K. Clayson, X. L. Pan, and J. Liu, "Regional variation of
corneal stromal deformation measured by high-frequency ultrasound elastography," (in
English), Experimental Biology and Medicine, vol. 246, no. 20, pp. 2184-2191, Oct 2021.
[127] D. A. Hoeltzel, P. Altman, K. Buzard, and K. I. Choe, "Strip Extensiometry for
Comparison of the Mechanical Response of Bovine, Rabbit, and Human Corneas," (in
English), Journal of Biomechanical Engineering-Transactions of the Asme, vol. 114, no.
2, pp. 202-215, May 1992.
[128] A. Elsheikh, B. Geraghty, P. Rama, M. Campanelli, and K. M. Meek,
"Characterization of age- related variation in corneal biomechanical properties," (in
English), Journal of the Royal Society Interface, vol. 7, no. 51, pp. 1475-1485, Oct 6 2010.
[129] D. Touboul et al., "Supersonic Shear Wave Elastography for the In Vivo Evaluation
of Transepithelial Corneal Collagen Cross-Linking," (in English), Investigative
Ophthalmology & Visual Science, vol. 55, no. 3, pp. 1976-1984, Mar 2014.
[130] K. Y. Zhang, X. Q. Qian, X. Mei, and Z. C. Liu, "An inverse method to determine
the mechanical properties of the iris in vivo," (in English), Biomedical Engineering Online,
vol. 13, May 30 2014.
[131] J. Heys and V. H. Barocas, "Mechanical characterization of the bovine iris," (in
English), Journal of Biomechanics, vol. 32, no. 9, pp. 999-1003, Sep 1999.
[132] X. Y. Zhang et al., "Noninvasive assessment of age-related stiffness of crystalline
lenses in a rabbit model using ultrasound elastography," (in English), Biomedical
Engineering Online, vol. 17, Jun 13 2018.
115

[133] C. Wu et al., "Assessing Age-Related Changes in the Biomechanical Properties of
Rabbit Lens Using a Coaligned Ultrasound and Optical Coherence Elastography System,"
(in English), Investigative Ophthalmology & Visual Science, vol. 56, no. 2, pp. 1292-1300,
Feb 2015.
[134] S. Park, H. Yoon, K. V. Larin, S. Y. Emelianov, and S. R. Aglyamov, "The impact
of intraocular pressure on elastic wave velocity estimates in the crystalline lens," (in
English), Physics in Medicine and Biology, vol. 62, no. 3, pp. N45-N57, Feb 7 2017.
[135] K. M. Myers, F. E. Cone, H. A. Quigley, S. Gelman, M. E. Pease, and T. D. Nguyen,
"The in vitro inflation response of mouse sclera," (in English), Experimental Eye Research,
vol. 91, no. 6, pp. 866-875, Dec 2010.
[136] I. Z. Nenadic, M. W. Urban, S. A. Mitchell, and J. F. Greenleaf, "Lamb wave
dispersion ultrasound vibrometry (LDUV) method for quantifying mechanical properties
of viscoelastic solids," (in English), Physics in Medicine and Biology, vol. 56, no. 7, pp.
2245-2264, Apr 7 2011.
[137] G. X. Lu et al., "Layer-specific ultrasound elastography using a multi-layered shear
wave dispersion model for assessing the viscoelastic properties," (in English), Physics in
Medicine and Biology, vol. 66, no. 3, Feb 7 2021.
[138] A. Elsheikh, B. Geraghty, D. Alhasso, J. Knappett, M. Campanelli, and P. Rama,
"Regional variation in the biomechanical properties of the human sclera," (in English),
Experimental Eye Research, vol. 90, no. 5, pp. 624-633, May 2010.
[139] I. A. Sigal, J. G. Flanagan, I. Tertinegg, and C. R. Ethier, "Finite element modeling
of optic nerve head biomechanics," (in English), Investigative Ophthalmology & Visual
Science, vol. 45, no. 12, pp. 4378-4387, Dec 2004.
[140] J. A. S. Rada, S. Shelton, and T. T. Norton, "The sclera and myopia," (in English),
Experimental Eye Research, vol. 82, no. 2, pp. 185-200, Feb 2006.
[141] J. M. Dunkin, A. V. Crum, R. S. Swanger, and S. A. J. Bokhari, "Globe Trauma,"
(in English), Seminars in Ultrasound Ct and Mri, vol. 32, no. 1, pp. 51-56, Feb 2011.
116

[142] N. Bayat et al., "A reversible thermoresponsive sealant for temporary closure of
ocular trauma," (in English), Science Translational Medicine, vol. 9, no. 419, Dec 6 2017.
[143] P. M. Pinsky, D. van der Heide, and D. Chernyak, "Computational modeling of
mechanical anisotropy in the cornea and sclera," (in English), Journal of Cataract and
Refractive Surgery, vol. 31, no. 1, pp. 136-145, Jan 2005.
[144] S. L. Y. Woo, W. A. Schlegel, A. S. Kobayashi, and C. Lawrence, "Nonlinear
Material Properties of Intact Cornea and Sclera," (in English), Experimental Eye Research,
vol. 14, no. 1, pp. 29-+, 1972.
[145] N. J. Jan et al., "Polarization microscopy for characterizing fiber orientation of
ocular tissues," (in English), Biomedical Optics Express, vol. 6, no. 12, pp. 4705-4718, Dec
1 2015.
[146] L. C. Ho et al., "Magic Angle-Enhanced MRI of Fibrous Microstructures in Sclera
and Cornea With and Without Intraocular Pressure Loading," (in English), Investigative
Ophthalmology & Visual Science, vol. 55, no. 9, pp. 5662-5672, Sep 2014.
[147] K. M. Myers, B. Coudrillier, B. L. Boyce, and T. D. Nguyen, "The inflation
response of the posterior bovine sclera," (in English), Acta Biomaterialia, vol. 6, no. 11,
pp. 4327-4335, Nov 2010.
[148] S. Nagase et al., "Anisotropic Alteration of Scleral Birefringence to Uniaxial
Mechanical Strain," (in English), Plos One, vol. 8, no. 3, Mar 11 2013.
[149] B. Coudrillier, C. Boote, H. A. Quigley, and T. D. Nguyen, "Scleral anisotropy and
its effects on the mechanical response of the optic nerve head," (in English), Biomechanics
and Modeling in Mechanobiology, vol. 12, no. 5, pp. 941-963, Oct 2013.

Abstract (if available)

Abstract Elastography is a widely used imaging modality to assess the biomechanical properties of tissue in a non-invasive manner, evidence shows that with the progression of certain diseases, the morphology changes are indiscernible, but the biomechanical changes are prominent, hence, elastography provides additional diagnosis information besides the conventional structural imaging.
Recently, elastography in ophthalmology is growing rapidly, for it can reconstruct the biomechanics non-invasively while maintain the original structure of the imaging object. Palpation, air puff, acoustic micro-tapping, acoustic radiation force (ARF) and shaker have been used to excite the ocular tissue and generate tissue motion, ultrasound and optical coherence tomography (OCT) have been used to track this tissue motion and reconstruct the ocular tissue biomechanics. However, palpation is limited by user-dependence, while air puff and acoustic micro-tapping suffer from the energy attenuation from tissue, those methods are not suitable for the posterior segment of the eye. Furthermore, most conventional ultrasonic elastography studies carry out a standard frequency range, which can not meet the need for high spatial resolution in ocular tissue.
The work presents in this dissertation proposal develops single element ultrasonic transducer based elastography, high frequency ultrasonic array based elastography, and optical coherence elastography (OCE) and investigates their applications in ophthalmology. One of the advantages that these systems have in common is the high resolution. This advantage enables to reconstruct the biomechanical properties of the imaging target accurately. In this study, the imaging or detecting method includes single element ultrasonic transducer, array transducer and OCT. Single element transducer based ultrasonic elastography is more capable of in vitro study. Comparing with the array transducer, the performance of the single element transducer is typically better owing to the manufactural difficulty of the array, for example, single element transducer gains better sensitivity and the bandwidth. However, single element transducer based ultrasonic elastography requires mechanical scanning during the measurement. The slow imaging speed restricts the application of this system into clinic, its high performance imaging transducer makes it more suitable for in vitro study. OCE has the highest resolution among these imaging systems, it also gains the advantage of the imaging the posterior eye due to the transparency of the cornea. It is capable of imaging one specific ocular tissue, but it can barely build the biomechanical connections with more ocular tissues due to the limited field of view.
Due to the regulation from U.S. Food and Drug Administration (FDA) on the ultrasound exposure in the eye, a shaker has been used instead of ARF to generate the tissue motion. After calibrated with imaging phantoms of our imaging systems, the biomechanics mappings of optic nerve head (ONH) and peripapillary sclera (PPS) have been provided by a single element transducer elastography system, then an advanced Lamb wave model has been applied with OCE system to reconstruct the biomechanical properties of ONH with more accuracy. Furthermore, the OCE system has been applied in vivo to reconstruct the biomechanics of the retina in living rabbits, and the biomechanical effects of the retinal prosthetic electrode on the retina have been further investigated. Finally, the quantitative biomechanics mappings of the in vivo rabbit eyes have been reconstructed under different intraocular pressure (IOP) levels with the array based elastography system. The anisotropy of the equatorial sclera has been revealed experimentally with the same system. These studies demonstrate the feasibility of high resolution elastography in the applications of ophthalmology.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

High frequency ultrasound elastography and its biomedical applications

PDF

2D ultrasonic transducer array’s design and fabrication with 3D printed interposer and applications

PDF

Configurable imaging platform for super-harmonic contrast-enhanced ultrasound imaging

PDF

High frequency ultrasonic phased array system and its applications

PDF

Multi-modality intravascular imaging by combined use of ultrasonic and opticial techniques

PDF

High-frequency ultrasonic transducers for photoacoustic applications

PDF

Intravascular imaging on high-frequency ultrasound combined with optical modalities

PDF

Development of high frequency focused transducers for single beam acoustic tweezers

PDF

Quantification of cellular properties using ultra-high frequency single-beam acoustic tweezer

PDF

High-frequency ultrasound array-based imaging system for biomedical applications

PDF

Ultrasound neuromodulation and its applications for noninvasive vision restoration

PDF

A high frequency array- based photoacoustic microscopy imaging system

PDF

Highly integrated 2D ultrasonic arrays and electronics for modular large apertures

PDF

Single-cell analysis with high frequency ultrasound

PDF

Additive manufacturing of piezoelectric and composite for biomedical application

PDF

Design and development of ultrasonic array transducers for specialized applications

PDF

Array transducers for high frequency ultrasound imaging

PDF

Fabrication of ultrasound transducer and 3D-prinitng ultrasonic device

PDF

Piezoelectric ultrasonic and acoustic microelectromechanical systems (MEMS) for biomedical, manipulation, and actuation applications

PDF

High frequency ultrasonic imaging for the development of breast biopsy needle with miniature ultrasonic transducer array

Asset Metadata

Creator Li, Runze (author)

Core Title High resolution elastography in ophthalmology

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Biomedical Engineering

Degree Conferral Date 2022-08

Publication Date 07/27/2022

Defense Date 06/02/2022

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

Tag elastic wave,OAI-PMH Harvest,Ophthalmology,optical coherence elastography,optical coherence tomography,ultrasound elastography,ultrasound imaging

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Zhou, Qifa (committee chair), Humayun, Mark (committee member), Shen, Keyue (committee member)

Creator Email runzeli@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC111375360

Unique identifier UC111375360

Legacy Identifier etd-LiRunze-11005

Document Type Dissertation

Format application/pdf (imt)

Rights Li, Runze

Type texts

Source 20220728-usctheses-batch-962 (batch), 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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

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

Repository Email cisadmin@lib.usc.edu