University of Southern California Dissertations and Theses

Design and fabrication of a high frequency PMN-PT needle transducer for retinal blood flow measurement

(USC Thesis Other)

Design and fabrication of a high frequency PMN-PT needle transducer for retinal blood flow measurement

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content NOTE TO USERS
This reproduction is the best copy available.
®
UMI
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DESIGN AND FABRICATION OF A HIGH FREQUENCY PMN-PT
NEEDLE TRANSDUCER FOR RETINAL BLOOD FLOW MEASUREMENT
Copyright 2005
by
Bruce Lai
A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(BIOMEDICAL ENGINEERING)
August 2005
Bruce Lai Student
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 1430395
Copyright 2005 by
Lai, Bruce
All rights reserved.
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, colored or poor quality illustrations and
photographs, print bleed-through, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
®
UMI
UMI Microform 1430395
Copyright 2006 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
I would like to acknowledge Dr. K. Kirk Shung for his help throughout the course of
this project. Thank you for providing a positive environment and the resources for
me to learn a great deal about ultrasound engineering. Thank you Manny for
teaching me about transducer design and fabrication. Without your direction, I
would have not been able to begin building the Doppler system. Thank you Sean for
designing and debugging the pulser/receiver board. Without it, I would have not
been able to obtain Doppler data. I would like to thank Jon Cannata for giving me
advice on many aspects of transducer design, fabrication, and testing. Your guidance
has been a great asset to me. Finally, I would like to thank Jay and Qifa for all of
your help in laboratory related matters.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iii
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ii
LIST OF TABLES vi
LIST OF FIGURES vii
ABSTRACT ix
Chapter 1 INTRODUCTION 1
1.0 Background 1
Chapter 2 ANATOMICAL APPLICATION 3
2.0 Introduction 3
2.1 Eye Anatomy 3
Chapter 3 FUNDAMENTALS OF DOPPLER TRANSDUCERS 6
3.0 Introduction 6
3.1 Piezoelectric Effect ' 7
3.2 Pulsed Wave vs. Continuous Wave 7
3.3 Transducer Properties 8
3.3.1 Half Wavelength Resonance 9
3.3.2 Sensitivity 10
3.3.3 Bandwidth 12
Chapter 4 PMN-PT PROPERTIES 13
4.0 Introduction 13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iv
4.1 Basic Properties 13
4.2 Single Crystal Structure 15
Chapter 5 DESIGN AND SIMULATION 16
5.0 Introduction 16
5.1 Aperture 16
5.2 Center Frequency 17
5.3 Modeling Parameters 18
5.3.1 Piezoelectric Material 18
5.3.2 Backing Layer 19
5.3.3 Matching Layers 19
5.4 Modeling Results 20
5.4.1 Modeled Input 20
5.4.2 Modeled Bandwidth 22
Chapter 6 FABRICATION AND TESTING 25
6.0 Introduction 25
6.1 Initial Fabrication 25
6.1.1 Piezoelectric Material 25
6.1.2 Matching Layers 26
6.1.3 Backing Layer 26
6.2 Final Assembly 29
6.2.1 Housing 29
6.2.2 Electrical Connection 30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V
6.3 Testing 32
6.3.1 Experimental Input Impedance 33
6.3.2 Experimental Bandwidth 36
6.3.3 Experimental Insertion Loss 41
Chapter 7 DOPPLER SYSTEM OVERVIEW 45
7.0 Introduction 45
7.1 Pulser and Receiver 45
7.2 Basic Processing 46
Chapter 8 PRELIMINARY DATA, CONCLUSIONS, AND FUTURE WORK 48
8.0 Introduction 48
8.1 Data Acquisition and Analysis 48
8.2 Conclusions 50
8.3 Future Work 51
BIBLIOGRAPHY 52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
vi
LIST OF TABLES
Table 3.1: Ideal impedances for one or two matching layers. 11
Table 4.1: Material properties for PMN-PT. 14
Table 5.1: PZCAD parameters for PMN-PT. 19
Table 6.1: Insertion loss voltage measurements. 43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
vii
LIST OF FIGURES
Figure 2.1: Overview of eye anatomy. 4
Figure 2.2: Doppler angle. 4
Figure 3.1: Basic transducer. 6
Figure 3.2: Doppler reflection. 9
Figure 5.1: Facial dimensions of transducer. 17
Figure 5.2: Simulated input impedance. 21
Figure 5.3: Simulated pulse echo. 22
Figure 5.4: Simulated bandwidth. 23
Figure 6.1: Fabrication of element plugs. 27
Figure 6.2: Final assembly of transducer. 30
Figure 6.3: Finished needle transducer. 33
Figure 6.4: Experimental input impedance magnitude. 34
Figure 6.5: Experimental input impedance phase. 34
Figure 6.6: PZT4 input impedance magnitude. 35
Figure 6.7: PZT4 input impedance phase. 36
Figure 6.8: Pulse echo equipment diagram. 37
Figure 6.9: Experimental pulse echo. 38
Figure 6.10: Experimental bandwidth. 38
Figure 6.11: PZT4 pulse echo. 40
Figure 6.12: PZT4 bandwidth. 40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
viii
Figure 6.13: Insertion loss equipment diagram. 42
Figure 6.14: Experimental insertion loss. 43
Figure 7.1: Doppler system overview. 46
Figure 7.2: Actual Doppler system. 47
Figure 8.1: Data acquisition location. 48
Figure 8.2: Spectrogram data. 49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ix
ABSTRACT
A 44 MHz Doppler needle transducer has been designed and fabricated for
measurement of retinal blood flow. The active element of the transducer is based on
lead magnesium niobate-lead titanate or PMN-PT.
Ideal PMN-PT properties are a high thickness mode coupling coefficient of 0.58 and
single crystal structure. Two quarter wavelength matching layers are utilized - a
silver particle epoxy at 7.33 MRayl acoustic impedance and a 2.59 MRayl parylene
coating. A conductive 5.9 MRayl epoxy serves as the backing. Measured electrical
input impedance is 74 Q, and actual -6 dB bandwidth is 38%. Accounting for
loading medium attenuation, insertion loss is 15.5 dB at 45 MHz.
Preliminary data show that the transducer is able to capture waveform data for
pulsatile blood flow. Frequencies range from 5 kHz to 15 kHz, corresponding to
maximum and minimum velocities of 28.9 cm/s and 9.6 cm/s respectively.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
Chapter 1
Introduction
1.0 Background
Diseases of the eye are commonly caused by age-related retinal vascular conditions,
where improper blood flow leads to retinopathy or death of light-sensing cells. One
possible scenario is wet macular degeneration, in which unchecked vascular growth
from the underlying choroid tissue layer invades the retina. These vessels can
rupture to cause hemorrhaging and retinal detachment, disrupting vision in the
central area known as the macula [1]. Estimates of macular degeneration prevalence
in people over 40 years of age currently stand at 1.47% [2]. On the other end of the
retinopathy spectrum is obstruction to blood flow. Retinal vein occlusions (RVOs)
are blockages in the vasculature that deprive the light-sensing cells of oxygen and
vital nutrients, thereby causing loss of vision. The underlying cause of such
occlusions has not yet been determined but may involve multiple factors including
abnormal blood constituents, atherosclerosis, and vessel anomalies [3]. Studies have
indicated that RVO frequency in those over the age of 49 is 1.6% [4]. Several
treatments have been developed or are in trial to combat RVOs, but no system has
been developed to gauge their efficacy in real time.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To fulfill this function, a Doppler system that is intended to measure blood flow in
the eye microvasculature has been designed and fabricated. As a critical part of this
system, a needle transducer or probe with high sensitivity has been developed. In
order to achieve good performance out of the small dimensions of the probe, a single
crystal material known as lead magnesium niobate-lead titanate (PMN-PT) is
employed in the active element of the transducer.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 2
Anatomical Application
3
2.0 Introduction
An understanding of basic anatomical features in the eye is necessary to describe the
application of the needle probe.
2.1 Eye anatomy
The function of the Doppler probe is to measure blood flow in the posterior region of
the eye where the retina resides. The transducer tip must be in close proximity to
retinal veins in order to obtain vascular data, which is not possible from outside the
eye. An incision in the outer layer of tissue called the sclera is to be made for deep
access. Major eye structures are displayed in figure 2.1 [5].
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
Transducer
Valve
Retina
Cornea
Sclera
Fig 2.1: Overview of eye anatomy and orientation of transducer with respect to the
retina [5].
Transducer
Transducer Axis
Target Vessel
Doppler Angle
Target Object
m
Figure 2.2: Doppler angle © between transducer tip and target vessel [6].
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The Doppler angle (0), the angle between the longitudinal axis of the transducer and
the blood flow [6], is estimated due to anatomical restrictions. The narrow field of
view from the front of the eye and the curvature of the retina impede highly accurate
angular measurement. Since 0 is an approximation, the resulting flow velocity
calculation according to equation 8.1 is also an approximation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 3
Fundamentals of Doppler Transducers
6
3.0 Introduction
A basic single element transducer is a device that consists of a backing layer, an
active piezoelectric material, one or two matching layers, and a housing. The face of
the transducer is known as the aperture.
Matching Layer
•Piezoelectric Material
Housing-
Backing
Electrical
Connector
Figure 3.1: Cross sectional view of a basic transducer
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
3.1 Piezoelectric Effect
The activity of the piezoelectric material comes from its polarized nature. Either
through voltage and temperature manipulation - poling - or through intrinsic
properties, the material is positively charged on one side and negatively charged on
the other. The transducer operates on the principle of the piezoelectric effect (i.e.
changing one form of energy to another). When the active material is stimulated by
an electrical pulse or multiple cycle bursts, its molecular structure allows it to
contract and expand to cause mechanical disturbances [7]. Such perturbations, when
directed properly, can be used to emit or pulse high frequency ultrasound waves
through a medium to an intended target. Reflections or echoes from the target can
then be detected and converted back to electrical form by the material.
3.2 Pulsed Wave vs. Continuous Wave
There are two main forms of Doppler ultrasound, pulsed wave (PW) and continuous
wave (CW). As its name implies, the CW approach involves a continuous sinusoidal
wave from the emitter. The output must be separated from the echoes to avoid
interference during later processing of the received signal. This requires that the
emitting and receiving elements be physically separate, leading to complicated
fabrication. Another drawback of CW is that continuous transmission and reception
of waves causes echoes to arrive from many different depths of tissue simultaneously,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
making Doppler signal localization impossible [8]. The main advantage of CW is
that each element can be optimized for its singular purpose.
A PW transducer is intended to send out a finite pulse train in a particular time
window during a cycle or repetition. Beyond this window, the device can receive
reflections without conflict. Additionally, a receiver gate can be implemented to
isolate certain depths of the beam path. Thus a single element can act as a pulser
and receiver, and transducer design is simplified. The transducer can be excited
multiple cycles per second in a particular pulse repetition frequency (PRF) [9].
Unable to accommodate separate pulsing and receiving elements, the Doppler needle
is of the PW type due to size restrictions.
3.3 Transducer Properties
In imaging ultrasound, the performance of a probe is determined by bandwidth and
sensitivity. A larger peak-to-peak echo is indicative of high sensitivity, which means
even targets that are poor reflectors can still be detected. A pulse that is short in the
time domain allows for high bandwidth and very fine resolution, meaning that small
targets in close proximity can be distinguished. Unlike imaging applications,
Doppler ultrasound relies more on sensitivity rather than bandwidth. Flow
information is carried in frequency changes instead of the actual echoes [10].
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Multiple voltage cycles are used to excite a Doppler transducer in order provide
suitable reflections for frequency shift and flow detection.
Transducer
Emitted Pulse
Reflection
Target Vessel
Target Object
in Flow
Figure 3.2: Doppler reflection due to moving targets in flow. The frequency of the
reflection is changed from that of the emitted pulse.
3.3.1 Half Wavelength Resonance
The transducer is designed to work at a specific frequency known as the center
frequency (fc ). When stimulated by a voltage pulse, the active element vibrates at a
particular wavelength, which is double the thickness of the piezoelectric material. At
1/2 thickness, the material is able to resonate at fc.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
3.3.2 Sensitivity
Since we are dealing with two different forms of energy, we need a way to quantify
their interconversion in the piezoelectric material. The thickness mode
electromechanical coupling coefficient (ECC) kt, and equations 3.1-3.4 are presented
for this purpose [11, 12]. Thickness mode - perpendicular to aperture plane - is set
by the shape and spatial orientation of the active material and is common for single
element transducers.
D = es +eS
T = cES -e E
S = dE + y ET
D = £TE + dT
(3.1)
(3.2)
(3.3)
(3.4)
ECC can theoretically vary from 0 (no piezoelectric activity) to 1 (perfect energy
transfer), but values closer to 0.5 are typical. A greater value of kt indicates larger
bandwidth and, more importantly, higher sensitivity [13]. In terms of the
piezoelectric equations, D is the dielectric displacement, C/m, T is stress, N/m, S is
strain, m, e is the clamped dielectric constant, F/m, c is the elastic stiffness, N/m , e
is the free dielectric constant, F/m, E is the electric field, V/m , e is the stress
• y > y
constant, C/m , d is the strain constant, C/N, and y is the elastic compliance, m /N.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
Piezoelectric material impedance, or resistance to energy transfer, affects device
sensitivity as well. On the electrical front, impedance magnitude should ideally be
matched to the 50 Q electronics that provide the voltage excitation, and impedance
phase should be as close to 0° as possible. However, the clamped dielectric constant,
C Q
6 3 3 , as well as the material dimensions determine this value. A small value of S 3 3 or
a small aperture indicates a low capacitance, which in turn leads to high impedance.
A large disparity in impedance can be mitigated through the use of electrical tuning.
In terms of acoustics, the impedance of the medium surrounding the probe is taken
into consideration. This loading medium is typically human tissue, which has an
acoustic impedance close to water - about 1.5 MRayl [14]. Acoustic impedance of
the piezoelectric material, which is the product of longitudinal sonic velocity through
the material and the material’s density, rarely matches that of tissue. Matching
layers of thickness XIA, whose impedances can be calculated through table 3.1, can
compensate for the difference [15].
Impedance First Layer (Zi) Second Layer (Zi)
One Layer Scenario Zcl,iZhm
Two Layer Scenario
r j 4/7 r j 3/7
A:
Z c 6 //Z l 1A/
Table 3.1: Ideal impedances for one or two matching layer scenarios. Zc is the
piezoelectric material impedance, and Zl is the loading medium impedance [15].
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
Matching both forms of impedance is critical in transducer behavior. By optimizing
the conversion of electrical to acoustic energy and vice versa, we are able to
maximize sensitivity.
3.3.3 Bandwidth
Like the ECC, the choice of backing material influences bandwidth, but ultimately
bandwidth is not a major concern with Doppler transducers. In imaging transducers,
heavy backings with high acoustic impedance may be necessary to absorb ringing
energy. If poorly damped, these mechanical oscillations result in long ultrasound
pulses and low resolution. The tradeoff for bandwidth is the loss of sensitivity. For
the Doppler needle, a light backing layer with optimized matching is the best
approach because sensitivity is the focus. Like the housing, the backing simply
serves as a mechanical support for the active material and as a conduction pathway
for electrical excitation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
Chapter 4
PMN-PT Properties
4.0 Introduction
To profile PMN-PT as an ideal choice for the Doppler needle, an overview of
material properties is presented. For comparison purposes, corresponding values for
other transducer materials are shown.
4.1 Basic Properties
The active element at the core of the Doppler needle is lead magnesium niobate-lead
titanate or PMN-PT. PMN-PT is a ceramic like some commonly used materials such
as lithium niobate (LNO) and lead zirconate titanate (PZT) but unlike the polymer
polyvinylidene fluoride (PVDF) [16,17]. Basic material properties are listed in table
4.1.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
PMN-PT PZT-4 LNO PVDF
Longitudinal Impedance
(MRayl)
36.86 34.50 34.06 4.31
Density (kg/m) 8000 7500 4640 1890
Longitudinal Velocity (m/s) 4608 4600 7340 2280
Clamped Dielectric
Constant
797 635 28 5
Clamped Capacitance (pF) 22.0 17.6 0.8 0.1
Coupling Coefficient 0.58 0.51 0.49 0.21
Table 4.1: Material properties for PMN-PT, lead zirconate titanate-4 (PZT-4),
lithium niobate (LN), and polyvinylidene fluoride (PVDF).
The thickness mode coupling coefficient and clamped dielectric constant for PMN-
PT make it ideal for the small dimensions of the needle transducer. The relatively
large value of kt promotes more efficient electromechanical energy conversion. A
clamped dielectric constant of 797 leads to a high clamped capacitance, which in
turn lowers electrical impedance for better matching to supporting pulser and
receiver electronics. The high acoustic impedance, however, is counterproductive to
sensitivity and must be overcome through matching layer utilization. This situation
is common to ceramics as they are very dense. The non-ceramic PVDF does not
suffer from acoustic mismatch but has signal loss due to poor electromechanical
coupling.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
4.2 Single Crystal Structure
The property of PMN-PT that is not listed in table 4.1 but contributes to optimal
piezoelectric behavior is that the material structure is composed of a single crystal.
All of its domains are lined up in one direction. Unlike polycrystalline PZT or LN,
PMN-PT does not have internal planes oriented in many different directions, which
allows for more directed mechanical energy and less internal dispersion.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
Chapter 5
Design and Simulation
5.0 Introduction
Transducer materials and material dimensions are chosen according to several
requirements. Factors such as sensitivity, durability, and small size are taken into
account in the design process.
5.1 Aperture
The needle transducer must meet certain criteria for its intraocular application. The
dimensions of the active element and its immediate housing must be able to slide
through an incision of less than 1 mm and reach the retina at the back of the eye. For
these considerations, a 20 gauge needle is chosen as the shell that holds the PMN-PT.
This particular size of needle has inner and outer diameters of 0.58 mm and 0.90 mm
respectively.
In order to fit inside the housing, the aperture cannot be larger than 0.58 mm in
diameter. Since it is difficult to cut a small circular face in crystalline PMN-PT
without any damage, a square 400 pm x 400 pm shape is chosen instead with a
diagonal length of 0.565 mm (Figure 5.1).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
Outer Needle Diameter
Inner Needle Diameter
Aperture
Figure 5.1: Relative facial dimensions of Doppler transducer.
5.2 Center Frequency
According to the equations 5.1 and 5.2 the small aperture of the probe (0.4 mm x 0.4
mm), approximated as a circle with radius r, mm, the center frequency fc , Hz, and the
speed of sound c (through PMN-PT), m/s, can be related to the wavelength X, pm,
and focal length Zf, mm [12]. Sonic speed in the piezoelectric material is
approximately 4600 m/s.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
A-j (5.1)
A sufficiently high center frequency must be selected because the facial dimensions
lead to a short focal depth. Multiple sinusoidal excitation cycles are to fit inside Zf.
However, a frequency that is too large presents fabrication difficulties due to half
wavelength resonance. The piezoelectric element can only be made so thin before
material integrity is compromised. A frequency of 45 MHz addresses these
conflicting demands and results in wavelength and focal length values of 102 pm and
1.57 mm respectively.
5.3 Modeling Parameters
From the preliminary dimension and frequency numbers, simulation in PiezoCAD
(PZCAD, Sonic Concepts, Bothell, WA) modeling software is possible. This
program is based on the KLM model that represents active elements as circuits.
5.3.1 Piezoelectric Material
A 45 MHz design center frequency is set as the first simulation parameter, and the
following piezoelectric element properties are entered.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
Material Thickness
(pm)
Longitudinal
Impedance
(MRayl)
Longitudinal
Velocity
(m/s)
Shear
Velocity
(m/s)
Density
(kg/m3 )
PMN-PT 51.2 36.864 4608 3258 8000
Clamped
Capacitance
(pF)
Clamped
Dielectric
Constant
Coupling
Coefficient
Electrical
Loss
Tangent
Mechanical
Loss
Tangent
22.0 797 0.583 0.0036 0.0036
Table 5.1: PZCAD parameters for PMN-PT material [18].
The thickness is 51 pm according to half-wavelength resonance. Element size is 400
pm x 400 pm as the chosen aperture.
5.3.2 Backing Layer
Since the bandwidth is not a critical issue, a conductive, low-impedance backing
material is needed. E-solder, with an impedance of 5.9 MRayl, meets these
requirements [19]. This substance is a silver based epoxy that can be mixed and cast
without much difficulty.
5.3.3 Matching Layers
On the other side of the active material, two matching layers are added into
PZCAD - one to act mainly as an acoustic transition between the PMN-PT and the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
other to serve as protection for the transducer tip. According to table 3.1 with Zc =
36.86 MRayl and Zl = 1.5 MRayl, the theoretical impedances for these layers are
9.35 MRayl and 2.36 MRayl. For the first layer, an established epoxy and silver
particle mixture is selected because of its mechanical strength and electrical
conductivity to the housing. The impedance is 7.33 MRayl, which is fairly close to
the ideal value while accounting for fabrication considerations [20]. The choice for
second layer is a parylene polymer, which is compatible with body tissues and seals
the probe face. This material has a 2.59 MRayl impedance. Quarter wavelength
thicknesses for the first and second layers are 11 pm and 13 pm respectively.
5.4 Modeling Results
The design parameters for PZCAD result in two primary simulated probe
characteristics - electrical input impedance and bandwidth. In practice, input
impedance or resistance is measured at the interface between the transducer and the
pulser/receiver and plotted over a frequency range. The bandwidth profile describes
transducer response at varying frequencies.
5.4.1 Modeled Input Impedance
The magnitude and phase modeling graphs for input impedance are in figure 5.2.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
230
Magnitude
Phase -45 220
210 -50
200 -55
S 190
-65 5 180
• * ■ <
I
170 -70 *
-75 160
-80 150
140 -85
-90 130
Frequency (MHz)
Figure 5.2: Electrical input impedance resulting from PZCAD simulation.
The phase line indicates capacitive or inductive losses due to the piezoelectric
material properties. A negative phase (capacitance) is common, and minimizing this
value helps to increase the Doppler device efficiency. At 42 MHz, the phase peaks
at -41°, which is close enough to 0 that tuning is not required. While this theoretical
center frequency is lower than the intended 45 MHz, it is still ideal for Doppler
applications. On the magnitude line, the corresponding location is halfway between
the series and parallel resonant frequencies (37 MHz and 47 MHz respectively).
These frequencies indicate points at which the transducer acts like KLM series and
parallel circuits. The impedance magnitude at 42 MHz is 180 Q, which is not
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
matched well to the 50 Q electronics. This is mainly caused by the tiny aperture and
its inverse proportionality to resistance. If necessary, the differential can be buffered
through proper cable lengths between the probe and electronics.
5.4.2 Modeled Bandwidth
Bandwidth results, while not critical to the performance of the Doppler needle, allow
further characterization of transducer behavior. Figures 5.3 and 5.4 show the pulse
echo and bandwidth response obtained from PZCAD.
Q .
-10
-15
0.05 0.15
Time (jxs)
0.2 0.25 0.3 0.35
Figure 5.3: Pulse echo or impulse response resulting from PZCAD simulation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10 20 30 40 50 60 70
Frequency (MHz)
Figure 5.4: Normalized bandwidth spectrum resulting from PZCAD simulation.
Ends of the -6 dB region are shown.
The main echo width is compact enough at 100 ns to be viable for the design
application. As a consequence of light backing utilization, a small amount of ringing
trails off after 120 ns, but overall performance should not be affected. The
bandwidth profile exhibits frequency response from 20 MHz to 70 MHz, where the -
6 dB bandwidth can be calculated according to equation 5.3.
BW % = x i oo% (5.3)
fC
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
Although the peak value is at 38 MHz (well under the design fc of 45 MHz), the 42
MHz frequency from the impedance analysis falls within the bandwidth of 57 % as
determined by using fc = 38 MHz, fi = 27 MHz, and f2 = 48 MHz. Sensitivity
should be adequate at the established resonant frequency.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
Chapter 6
Fabrication and Testing
6.0 Introduction
Fabrication is divided into two primary sections, making element plugs and
assembling the actual transducer. The finished Doppler probe is then subjected to
testing to gauge performance. Important data obtained are electrical input
impedance, bandwidth, and insertion loss.
6.1 Initial Fabrication
The first phase of fabrication involves active element manufacturing. Each element
is a plug consisting of PMN-PT, the first matching layer, and a backing layer.
Materials are temporarily wax-bonded to glass plates at each step to ensure a flat and
stable foundation for curing. Where electrodes are needed, a 500 A chrome layer
followed by a 1000 A gold layer are deposited through sputtering.
6.1.1 Piezoelectric Material
Fabrication of the Doppler needle begins with a poled sample of PMN-PT (HC
Materials Corp., Urbana, IL). The material is manufactured at approximately 1 cm x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
1 cm in the aperture plane and 700 pm in thickness. The positive and negative faces
have been sputtered with gold electrodes. Lapping is required to reach the 51 pm
design thickness (Figure 6.1a). In the process, the negative electrode is worn away,
and a new one is sputtered on as a replacement.
6.1.2 Matching Layers
A thin layer of Chemlok AP-131 (Lord Corp., Cary, NC) adhesion promoter is
swabbed on the negative face of the PMN-PT to help bond the first matching layer to
the negative PMN-PT face. For the matching layer material, and epoxy base is first
created, which is a mixture of 5 g Insulcast 501 and 0.65 g Insulcure 9 (American
Safety Technologies, Roseland, NJ). 1.25 g of the epoxy is combined with 3 g of
silver particles (Sigma-Aldrich Inc., St. Louis, MO) and scooped into a ceramic ring
that surrounds the piezoelectric material. Centrifugation is done at 3000 rpm for 15
minutes to ensure good contact between matching layer and PMN-PT (Figure 6.1b).
The silver particle epoxy is allowed to cure overnight at room temperature and
subsequently lapped down to 11 pm (Figure 6.1c).
6.1.3 Backing Layer
For the backing layer, a conductive epoxy called E-solder (Vonroll Isola, New
Haven, CT) is utilized. The active material-matching layer stack is flipped over to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
expose the positive electrode, and adhesion promoter is laid down again. E-solder is
mixed and poured into a ceramic ring over the PMN-PT. The backing is centrifuged
and cured with the same method as the first matching layer (Figure 6.Id) and lapped
down to under 3 mm in thickness (Figure 6.1e).
PMN-PT
:: Glass Plate
Silver Particle
Matching Layer
PMN-PT
Ceramic Ring
: Glass Plate
(b)
Figure 6.1: Fabrication steps to make element plugs. Cross sections are shown, (a)
Lapping PMN-PT after fixation to glass plate with wax. (b) Casting and curing
silver particle matching layer in machinable ceramic ring.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
Silver Particle
Matching Layer Ceramic Ring
Glass Plate
Silver Particle
Matching Layer
Ceramic Ring PMN-PT
Conductive Epoxy
Backing
Glass Plate
Silver Particle
Matching Layer
Ceramic Ring
Conductive Epoxy
Backing
GlassPlate
(e)
Figure 6.1: Continued, (c) Lapping matching layer, (d) Flipping over PMN-PT and
matching layer and fixing again with wax. Casting and curing conductive epoxy (E-
solder) backing in another ceramic ring, (e) Lapping backing layer.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
6.2 Final Assembly
Assembly of the Doppler transducer involves the supporting materials for the active
element. The element is fixed inside the head of a needle housing and wired up to an
SMA connector. These fabrication steps allow the probe to be interfaced to
pulser/receiver electronics.
6.2.1 Housing
Element plugs are cut out at 400 pm x 400 pm aperture with a dicing saw, where one
is used per transducer. A plug is inserted into an electrically insulating tube of
polyimide (inner diameter of 574 pm and outer diameter of 624 pm) until the silver
matching layer is lined up with the end of the tube. To fix the element in place and
to connect a positive lead wire to the backing, E-solder is carefully injected through
the back end of the polymer tube and cured but not centrifuged (Figure 6.2a).
A simple stainless steel housing is made by melting the plastic off of a 1.5 inch long
20 gauge needle and cleaning it in an ultrasonic bath. The element polyimide piece
is inserted into the housing, and Epotek 301 epoxy (Epoxy Technology, Billerica,
MA) is wicked in through facial gaps. After an overnight room temperature curing
period, excess Epotek is carefully removed by sanding and acetone swabbing (Figure
6.2b).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
6.2.2 Electrical Connection
The face of the active element is sputtered (500 A chrome and 1000 A gold) to form
an electrical connection between the element and ground plane. The entire length of
the needle is coated with a 13 pm layer of parylene through vapor deposition (Figure
6.2c). A larger threaded housing lathed out of brass is attached to the back of the
needle with E-solder. The exposed end of the positive lead wire is soldered into an
SMA connector and secured with Epotek. Curing is again performed. The
connector is screwed into the threaded portion of the housing and fixed with E-solder
(Figure 6.2d).
Element
Plug
Polyimide
Tubing
E-solder
Positive Lead
Wire
(a)
Figure 6.2: Fabrication steps to assemble transducer from active element. Cross
sections are shown. Subfigures a to c are magnifications of the transducer tip. (a) E-
solder fixation in polyimide tubing with positive lead wire attachment. Individual
layers are not shown in the element.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
Needle
Housing
Element
Plug F > > > t* > 2
Polyimide
Tubing
E-solder
t\w,v
Positive Lead
Wire
(b)
Element
Plug
E-solder
Parylene
Coating
Needle
Housing
Polyimide
Tubing
Positive Lead
Wire
(c)
Figure 6.2: Continued, (b) Insertion into needle housing and filling facial gaps with
epoxy, (c) Parylene vapor deposition.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Parylene Coated
Needle Housing
Positive Lead
Wire
Rear Brass
Housing
SMA Connector
(d)
Figure 6.2: Continued, (d) SMA connector attached to larger rear housing.
6.3 Testing
The finished Doppler transducer (Figure 6.3) is tested to quantify its real
performance, which is compared to the PZCAD simulation. Electrical input
impedance, bandwidth, and insertion loss are measured. Data from a PZT4
transducer (manufactured using the fabrication techniques outlined in this chapter)
are included for relative characterization of different piezoelectric materials.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
Figure 6.3: Finished Doppler needle transducer.
6.3.1 Experimental Input Impedance
Input impedance is measured through the Z-probe function of a Yokogawa Hewlett-
Packard 4194A impedance analyzer (Hewlett-Packard Japan, Ltd., Tokyo, Japan)
Impedance magnitude and phase follow in figures 6.4 and 6.5. PZT4 plots are
integrated into figures 6.6 and 6.7.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
2 4 0
Experimental
Simulated 220
200
180
Q , 160
a 140
120
100
60
40 45 5(
Frequency (MHz)
60 30 35
Figure 6.4: Experimental input impedance magnitude compared to simulated input
impedance magnitude.
-40
Experimental
Simulated -45
-50
-55
-65
o- .70
-75
-80
-85
-90
50
(MHz)
60 30 40
Frequency (MHz)
Figure 6.5: Experimental input impedance phase compared to simulated input
impedance phase.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
Compared to the simulated value of 180 ft at 42 MHz, the actual impedance of 74 ft
at 44 MHz is closer to the ideal 50 ft and 45 MHz parameters. This improved match
to the support electronics and proximity to the intended center frequency should help
sensitivity. Although the -58° phase (44 MHz) is slightly larger than the modeled -
40° value (42 MHz), capacitive losses should not pose a major problem. The higher
fc and lower impedance could be due to small deviations in fabrication. If the PMN-
PT were lapped a few microns under the design thickness, a shorter half wavelength
resonance and resistance path would result.
240
PMN-PT
PZT4 220
200
180
§ 160
< w
T3
3 140
a
I 120
100
60
Frequency (MHz)
Figure 6.6: PMN-PT needle impedance magnitude compared to that of a PZT4
transducer.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
-5 5
PMN-PT
PZT4 -60
-65
-70
-75
o o
-80
fc -85
-90
-95
-100
-105
60 40 45 50
Frequency (MHz)
Figure 6.7: PMN-PT needle impedance phase compared to that of a PZT4 transducer.
The phase plot for the PZT4 device peaks at 45 MHz, which corresponds to a
magnitude and phase of 80 Q and -62° respectively. These values are slightly higher
than the Doppler needle counterparts of 74 Q and -58°. A probe utilizing PMN-PT is
better matched to the 50 Q pulser/receiver hardware and suffers less from negative
phase capacitance.
6.3.2 Experimental Bandwidth
The pulse echo and bandwidth are measured using a water bath and quartz target.
Quartz is a strong reflector, thus minimizing energy losses from transmission
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
through the target. Distance between the transducer aperture and the reflector is the
focal length (1.57 mm). The pulser and receiver are combined into a Panametrics
PR5900 (R/D Tech Instruments Inc., Waltham, MA) unit set in P/E mode with 1 pJ
output energy, 1 KHz pulse repetition frequency, 50 C l coupling, band pass filtering
between 10 and 100 MHz, 0 dB receiver gain, and 0° output phase. Data is
visualized on a LeCroy LC534 (LeCroy Corp., Chestnut Ridge, NY) digital
oscilloscope (Figure 6.8). Resulting averaged plots are in figures 6.9 and 6.10. The
reflection is initially measured in millivolts, but the PZCAD counterpart is displayed
in units of millivolts per volt of pulser excitation. Obtained from the PR5900
specifications, a 100 V input is factored into the experimental data. PZT4 pulse echo
and bandwidth data are shown in figures 6.11 and 6.12.
Oscilloscope
4 oO Q Cables
Pulser/Receiver
Transducer
Quartz Target
Water Bath
Figure 6.8: Diagram of equipment used to obtain pulse echo and bandwidth data.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
2 0
Experimental
Simulated
-10
-15
0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Time (p.s)
Figure 6.9: Experimental pulse echo compared to simulated pulse echo.
Experimental
Simulated
-10
-15
< o
•o
a
& -20
-25
-30
-35
30 40 50 60 70
Frequency (MHz)
Figure 6.10: Experimental bandwidth spectrum compared to simulated bandwidth
spectrum. Both plots are normalized. Ends of the -6 dB region are shown.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
The actual echo waveform is smaller in peak-to-peak amplitude than the modeled
one. Some of this discrepancy is account for by the lack of medium attenuation and
scattering in the simulation. Water absorbs acoustic energy, preventing it from
returning to the transducer. In terms of bandwidth, the maximal response frequency
of 42 MHz is higher than that of the PZCAD spectrum and is closer to the 45 MHz
design frequency. As in the impedance measurement, slightly thinner piezoelectric
material could be responsible for this result. The -6 dB bandwidth of 38% (fi = 31
MHz and £2 = 47 MHz) is much lower than 57% predicted by the modeling software.
The higher center frequency accounts in part for the smaller percentage. However,
the experimental bandwidth profile is narrower than expected. This could be due to
imperfect bonding between the various layers of the probe tip, which causes internal
acoustic reflections and ringing in the output pulse. Nevertheless, 38% bandwidth is
sufficient for Doppler applications.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
PMN-PT
PZT4
CL,
-10
0.35 0.4 0.2 0.25
Time (|xs)
0.3 0.05
Figure 6.11: PMN-PT needle pulse echo compared to that of a PZT4 transducer.
PMN-PT
PZT4
-10
-20
-25
-30
60 35 40
Frequency (MHz)
20 30
Figure 6.12: PMN-PT needle bandwidth spectrum compared to that of a PZT4
transducer. Both plots are normalized. Ends of the -6 dB region are shown.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
The echo of the Doppler probe is much larger than that of the PZT4 transducer, a
discrepancy that is indicative of the needle transducer’s stronger signal performance.
With a center frequency of 45 MHz, the -6 dB bandwidth of the PZT4 device is 42%
(fi = 33 MHz, fi = 52 MHz), a number that is marginally higher than that of the
needle transducer. The bandwidth performances of PMN-PT and PZT4 are similar.
6.3.3 Experimental Insertion Loss
Insertion loss allows for quantification of sensitivity, which is the primary focus of
the needle transducer design. A similar water bath setup as the bandwidth test is
used to measure insertion loss, with the difference being a larger separation of 3 mm
between the target and aperture. A Sony Tektronix AFG2020 function generator
(Tektronix Inc., Beaverton, OR) set to output 10 sinusoidal cycles at 4 V connected
to the oscilloscope at 50 Q coupling. The actual amplitude is found to be 4.01 V.
Then the probe is connected to both the function generator and the oscilloscope, and
the coupling is changed to 1 MO. The transducer is positioned approximately 4.5
mm from the quartz target (Figure 6.13).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
Oscilloscope
Transducer
'50 Q Cables
’ unction Generator
Quartz Target
/ Water Bath
Figure 6.13: Diagram of equipment used to obtain insertion loss data.
Reflection amplitudes are measured in a 20 MHz range centered at 43 MHz.
Averaged data is shown in table 6.1, and the relevant insertion loss plots is in figure
6.14. Insertion loss calculations are performed according to equation 6.1, where Vi
is the voltage emitted by the function generator (4.01 V), V0 is the received voltage
from the reflection, and IL is the insertion loss. Water attenuation is accounted for
by a factor of 2.2 x 10^ dB/mm-MHz2.
IL = -20 x log —
. V ,
(6.1)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
Frequency (MHz) Reflection
Amplitude (mV)
Frequency (MHz) Reflection
Amplitude (mV)
30 109.5 43 426
31 124.3 44 512
32 149.5 45 533
33 174.1 46 527
34 198.8 47 478
35 211.3 48 412
36 208.8 49 327
37 188.0 50 226
38 159.7 51 145
39 141.4 52 89.5
40 156.4 53 57.0
41 223.8 54 38.7
42 323.8 55 25.0
Table 6.1: Various insertion loss reflection voltage measurements and corresponding
frequencies.
45
40
35
30
25
20
15
35 40 45 50 55 30
Frequency (MHz)
Figure 6.14: Insertion loss plot from 30 MHz to 55 MHz. Values are calculated from
equation 6.2 and table 6.1. Water attenuation is taken into consideration.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
As indicated by figure 6.14, there is a minimal insertion loss of 15.5 dB at 45 MHz.
While not an extremely low value, it is acceptable for strong signal performance.
Even weak reflectors such as tissue and blood cells intended for the Doppler needle
should provide detectable echoes for processing and flow analysis.
Insertion loss for the PZT4 transducer is so high that only the maximal reflection
voltage of 92.3 mV at 49 MHz is recorded. The function generator outputs 3.83 V
during this round of testing. Applying equation 6.1 and the water attenuation factor
to these data, one can see that the loss for PZT4 is 30.8 dB. The contrast with the
much lower Doppler probe value demonstrates that PMN-PT has superior sensitivity
due to its higher coupling coefficient and dielectric constant.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 7
Doppler System Overview
45
7.0 Introduction
A custom Doppler system is designed to support the transducer in its flow
measurement function. The main components are pulser/receiver electronics with
basic signal processing and a computer with an analog-to-digital (A/D) converter and
data capture software.
7.1 Pulser and Receiver
The main components of the pulser/receiver circuitry are a 45 MHz crystal oscillator,
a timing circuit, a power amplifier, wide-band low-noise amplifiers, quadrature
coherent demodulator, sample-and-hold circuitries, and audio amplification. The
pulser portion provides an 80 V peak-to-peak 10-cycle sinusoidal excitation train for
the Doppler transducer by windowing the crystal output with the timing circuit and
amplifying it with the power amplifier. Using an external amplifier and band pass
filter, the receiver section accepts echoes detected by the probe, magnifies their
amplitude, and filters them to reduce electronic noise.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
7.2 Basic Processing
Demodulation is performed on the receive signal separately from the board to extract
the Doppler information and to mitigate noise effects. Further processing is done
through low pass filtering and audio amplification, and the data is fed into a
computer through a sound card that acts as an A/D converter. PRF interference is
filtered out as well. Waveform capture in sound file format is done by Spectrogram
audio analysis software. A diagram of the Doppler system is shown in figure 7.1.
Reference Signal
i r
45 MHz Burst Frequency Data
Doppler Signal
Pulse
’ ransducer
Echo
Audio Out
Sound Card
(A/D Converter)
Low Pass
Filter
Audio Amplifier
Demodulator Power Amplifier
Data Processing
Sample and Hold,
PRF Filter
Timing Circuitry
Diode Limiter,
Amplifier, BP Filter
45 MHz Oscillator
Spectrogram
Display and Capture
Figure 7.1: Doppler system overview.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
Figure 7.2: Doppler system. The needle transducer is located near the center of the
image.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
Chapter 8
Preliminary Doppler Data,
Conclusions, and Future Work
8.0 Introduction
An initial sample of data is acquired from the wrist vasculature to determine the
Doppler system functionality. Further tests are required for detailed characterization
of primarily the needle transducer and secondarily the support electronics.
8.1 Data Acquisition and Analysis
The preliminary data acquisition protocol is non-invasive and can be performed
relatively quickly. The Doppler data capture method is displayed in figure 8.1.
Figure 8.1: Location of data acquisition for the Doppler needle in the wrist area.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
. Time (sec)
Figure 8.2: Spectrogram data sample.
Background noise is evident in the spectrogram, which is mainly due to electronic
interference from the oscillator crystal in the ground plane of the pulser/receiver
electronics. Nevertheless, the waveform profile is still distinguishable. Peak and
trough frequency values can be used to estimate pulsatile flow velocities when
applied to equation 8.1 [6].
v = — ^ — (8.1)
2 f c cos©
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
Where fd is the Doppler frequency, fc is the resonant frequency of the transducer (45
MHz), c is the speed of sound in tissue (1500 m/s), and 0 is the estimated Doppler
angle. Maximum and minimum frequencies from the Doppler data average 15 kHz
and 5 kHz respectively. Assuming an angle of about 30°, the corresponding flow
speeds are 28.9 cm/s and 9.6 cm/s.
8.2 Conclusions
A pulsed wave Doppler system based on a needle probe with a design center
frequency 45 MHz has been shown to be feasible. Although the actual system
operated around 44 MHz, this discrepancy was negligible in testing and application.
Choice of PMN-PT as the piezoelectric material and less emphasis on bandwidth
considerations allowed for relatively simple design. In particular, the uniform crystal
structure and high coupling coefficient of the PMN-PT promoted high sensitivity,
making it suitable for single element Doppler devices. Fabrication, while time
consuming and prone to small deviations, was ideal for creation of multiple probes
due to the small aperture size.
Overall, the sensitivity aspects of PZCAD simulation were similar to those of testing.
The high signal strength of the needle transducer was exhibited in the overlap
between the PZCAD and measured pulse echoes. The insertion loss was not high
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
enough to cause major performance degradation. Actual impedance was better than
that of the model, so electrical tuning was not necessary.
When used in conjunction with the custom pulser and receiver electronics, the needle
transducer was sensitive enough to measure small vessel flow in the extremities.
Despite oscillator noise, frequency data was still visible in the spectrogram data.
8.3 Future work
Further refinement of the supporting pulser/receiver electronics can be implemented
in subsequent revisions of the Doppler system. By using transformers, ground
planes can be isolated from the oscillator, leading to a lower noise floor and
enhanced transducer performance.
Computer data analysis can be improved with real time flow velocity measurements.
Quadrature demodulation can also implemented to obtain phase data for flow
direction determination. These refinements will contribute to a streamlined system
that can reliably and easily capture data in future ophthalmological studies.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
BIBLIOGRAPHY
[12] D. Berlincourt. “Piezoelectric Crystals and Ceramics.” in Ultrasonic
Transducer Materials, O.E. Mattiat, Ed., New York: Plenum Press, 1971.
[13] C.S. Desilets, J.D. Fraser, and G.S. Kino. “Design of Efficient Broad-Band
Piezoelectric Transducers.” IEEE Transactions on Sonics and Ultrasonics, vol.
SU-25, no. 3, pp. 115-125, May 1978.
[2] D.S. Friedman. “Prevalence of age-related macular degeneration in the United
States.” Archives o f Ophthamology, vol. 122, pp. 564-572, Apr. 2004.
[8] GE Healthcare. “Continuous wave Doppler.” Internet:
http://www.amershamhealth.com/medcvclopaedia/medicaWolume%20I/CQN
TINUOUS%20WAVE%20%20CW%20%20DQPPLER.asp. 2004 [June 1,
2005].
[6] GE Healthcare. “Doppler angle.” Internet:
httn://www.amershamhealth.com/medcvclopaedia/medicaWolume%20I/DQP
PLER%20ANGLE.asp. 2004 [June 1,2005].
[9] GE Healthcare. “Pulsed wave Doppler.” Internet:
http://www.amershamhealth.com/medcvclopaedia/medical/Volume%20I/pulse
d%20Doppler%20ultrasound.asp. 2004 [June 1,2005].
[18] HC Materials Corporation. “PMN-PT properties.” Internet:
http://www.hcmat.com/Pmn_Properties.html, [2005 Jun 1]
[10] R.K. Hung, P. Zimmerman, A. Duerinckx, M. Mellany, and E. Grant.
“Doppler.” Internet: http://ei.rsna.org/ei3/0079-98.fin/doppler.htm. June 21,
1999 [June 1, 2005].
[4] P. Mitchell, W. Smith, and A. Chang. “Prevalence and associations of retinal
vein occlusion in Australia.” Archives o f Ophthamology, vol. 114, pp. 1243-
1247, Oct. 1996.
[20] T.A. Ritter, K.K. Shung, R.L. Tutwiler, and T.R. Shrout. "Medical imaging
arrays for frequencies above 25 MHz," in Proceedings o f the IEEE Ultrasonics
Symposium, 1999, pp. 1203-1208.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
[17] M.D. Sherar and F.S. Foster. “The design and fabrication of high frequency
poly(vinylidene fluoride) transducers.” Ultrasonic Imaging, vol. 11, pp. 75-94,
Oct. 1989.
[14] K.K. Shung, M.B. Smith, and B. Tsui. Principles of Medical Imaging. New
York: Academic Press, 1992, pp. 78-161.
[15] K.K. Shung and M.J. Zipparo. “Ultrasonic transducers and arrays.” IEEE
Engineering in Medicine and Biology, vol. 15, pp. 20-30, Nov./Dec. 1996.
[I] J.W. Sowka, A.S. Gurwood, and A.G. Kabat. “Macular degeneration.” Internet:
http://www.revoptom.com/handbook/SECT5H.HTM, 2001 [June 1,2005].
[3] J.W. Sowka, A.S. Gurwood, and A.G. Kabat. “Macular degeneration.” Internet:
http://www.revoptom.com/handbook/sect5f.htm, 2001 [June 1,2005].
[19] H. Wang, T.R. Ritter, W. Cao, and K.K. Shung. “Passive Materials for High
Frequency Ultrasonic Transducers,” in Proceedings o f the SPIE Medical
Imaging, Ultrasonic Transducer Imaging, 1999, pp.35-42.
[7] Valpey Fisher Corporation. The User’ s Guide to Ultrasound and Optical
Products. Hopkinton, MA: Valpey-Fisher, 1996.
[5] K.M. Van De Graff. Human Anatomy. New York: McGraw Hill, 2002, p. 512.
[II] M.J. Zipparo. “Very high frequency (50 to 100 MHz) ultrasonic transducers
for medical imaging applications.” M.S. thesis, Pennsylvania State University,
University Park, PA, 1996.
[16] M.J. Zipparo, K.K. Shung, and T.R. Shrout. “Piezoceramics for high-
Frequency (20 to 100 MHz) single-element imaging transducers.” IEEE
Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, vol. 44,
pp. 1038-1048, Sep. 1997.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Development of high frequency annular array ultrasound transducers

PDF

Design of a portable infrared spectrometer: application to the noninvasive measurement of glucose

PDF

Evaluation of R.F. transmitters for optimized operation of muscle stimulating implants

PDF

A finite element model of the forefoot region of ankle foot orthoses fabricated with advanced composite materials

PDF

Initiation of apoptosis by application of high-intensity electric fields

PDF

Cellular kinetic models of the antiviral agent (R)-9-(2-phosphonylmethoxypropyl)adenine (PMPA)

PDF

Computation and validation of circulating blood volume with the indocyanine green dilution technique

PDF

A multimodal screen reader for the visually impaired

PDF

Characteristics and properties of modified gelatin cross-linked with saline for tissue engineering applications

PDF

A fluorescence microscopy study of quantum dots as fluorescent probes for brain tumor diagnosis

PDF

Effects of prenatal cocaine exposure in quantitative sleep measures in infants

PDF

An open ear canal sound delivery system

PDF

Comparisons of deconvolution algorithms in pharmacokinetic analysis

PDF

Head injury biomechanics: Quantification of head injury measures in rear-end motor vehicle collisions

PDF

Changes in medial gastrocnemius architecture with spasticity and contracture

PDF

Contact pressures in the distal radioulnar joint as a function of radial malunion

PDF

Functional impacts of morphology on synaptic transmission

PDF

A case study comparing measured and simulated acoustical environments: Acoustical improvements of Verle Annis Gallery

PDF

Biological materials investigation by atomic force microscope (AFM)

PDF

Hebel design analysis program

Asset Metadata

Creator Lai, Bruce (author)

Core Title Design and fabrication of a high frequency PMN-PT needle transducer for retinal blood flow measurement

School Graduate School

Degree Master of Science

Degree Program Biomedical Engineering

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

Tag engineering, biomedical,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Shung, K. Kirk (committee chair)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-45985

Unique identifier UC11337866

Identifier 1430395.pdf (filename),usctheses-c16-45985 (legacy record id)

Legacy Identifier 1430395.pdf

Dmrecord 45985

Document Type Thesis

Rights Lai, Bruce

Type texts

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

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

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA