810 IEEE TRANSACTIONS ON ULTRASONICS, FERROELECTRICS, AND FREQUENCY CONTROL, VOL. 67, NO.
4, APRIL 2020
Single-Crystal High-Frequency Intravascular
Ultrasound Transducer With
40-μm Axial Resolution
Xiaowen Ma and Wenwu Cao
Abstract — Intravascular ultrasound (IVUS) is one of the
most useful tools available today to assist intravascular
stenting procedures. Having higher resolution is very impor-
tant for helping doctors to evaluate the nature of atheroscle-
rotic plaques. The current commercial IVUS systems have
a spatial resolution of 70–200 µm in the axial direction and
200–400 µm in the lateral direction, which are insufficient for
accurate diagnosis. We report here a three-matching-layer
IVUS transducer design using a 0.72Pb(Mg1/3Nb2/3 )O3 −
0.28PbTiO3 single crystal, which can improve the axial
resolution to 40 µm without sacrificing the penetration
depth. Wire phantom imaging and in vitro porcine coronary
artery imaging show noticeably better axial resolution and
similar penetration depth compared with a commercial IVUS
transducer.
Index Terms — Broad bandwidth, high-frequency
transducer, intravascular ultrasound (IVUS),
three-matching-layer.
I. I NTRODUCTION
C ARDIOVASCULAR diseases are the leading cause of
human death globally. According to WHO’s report, more
than 17.9 million deaths per year occurred in 2015 due to
cardiovascular diseases alone, and the death toll is expected
to increase to more than 23.6 million by 2030. In the United
States, coronary heart disease is the leading cause (43.8%)
of death among cardiovascular diseases, followed by stroke
(16.8%), heart failure (9%), high blood pressure (9.4%), dis-
eases of the arteries (3.1%), and other cardiovascular diseases
(17.9%) [1]. Atherosclerosis is the main cause of coronary
heart disease. Effective diagnosis of atherosclerotic coronary
disease by identifying and quantifying atherosclerotic lesions
has been one of the main clinical issues [2]. The most
effective method is intravascular ultrasound (IVUS), which has
been used for many years in clinical applications to visualize
the inner structure of coronary arteries and detect high-risk
coronary plaque characteristics [3], [4].
The currently available commercial IVUS catheters include
a high-frequency ultrasound transducer, either a rotating
Manuscript received July 29, 2019; accepted November 21, 2019. Date
of publication November 28, 2019; date of current version March 25,
2020. This work was supported by the Shanghai Science and Tech-
nology Committee through the Shanghai Sailing Program under Grant
17YF1405300. (Corresponding author: Wenwu Cao.)
X. Ma is with the School of Life Science and Technology, Harbin Institute
of Technology, Harbin 150080, China.
W. Cao is with the School of Life Science and Technology, Harbin
Institute of Technology, Harbin 150080, China, and also with the Depart-
ment of Mathematics and Materials Research Institute, The Pennsylvania
State University, University Park, PA 16802 USA ( e-mail: dzk@psu.edu).
. .
Digital Object Identifier 10.1109/TUFFC.2019.2956603
0885-3010 © 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
. .
Authorized licensed use limited to: Hong Kong University of Science and Technology. Downloaded on August 14,2020 at 15:14:21 UTC from IEEE Xplore. Restrictions apply.
single-element transducer or an array transducer, with their
center frequency (FC) ranging from 20 to 40 MHz. Their
spatial resolution is usually 70–200 μm in the axial direction
and 200–400 μm in the lateral direction [5]–[7]. In order
to improve the accuracy of clinical diagnosis with higher
spatial resolution, many trial designs have been explored,
such as single-crystal, composite transducers referred to as
pc-MUT [8], ultrahigh-frequency IVUS [9], optical coher-
ence tomography (OCT)-integrated IVUS [10], and the newly
released 60-MHz composite transducer-based IVUS catheter
(OPTICROSS HD, Boston Scientific). However, the pc-MUT
and ultrahigh-frequency transducer technologies introduce sig-
nificant difficulties in the fabrication process that lower the
manufacturing yield. For example, the kerf edge straightness
of pc-MUT composite is difficult to be ensured during the deep
dry etching process. For ultrahigh-frequency (over 80 MHz)
IVUS transducers, the piezocrystal needs to be lapped down
to a thickness of less than 30 μm, which is difficult to achieve
[11]. For IVUS-OCT dual-modality catheters, several designs
have demonstrated the feasibility of this technology, which
allows more accurate plaque characterization than using IVUS
or OCT alone [12]–[14]. However, some technical and cost
issues still remain to be solved for this complex hybrid system,
such as increasing the imaging speed and the coregistration
accuracy between IVUS and OCT [15]. The new commercial
60-MHz IVUS catheter is made of 1–3 piezocomposites; it
is very difficult to use the low-cost dice-and-fill technology
due to the strict kerf width requirement (less than 4 μm)
[16] at such high frequencies. Moreover, the penetration depth
of the 60-MHz transducer is inferior to the low-frequency
(40 MHz) transducer, so the increased spatial resolution is at
the cost of shallower penetration depth. Besides the technolo-
gies mentioned above, dual-frequency [17] or multi-frequency
[18] IVUS transducers were also designed for enhancing
the spatial resolution while retaining satisfactory penetration
depth. However, the accurate alignment of dual/multiple trans-
ducers is very critical for image co-registration. Ma et al.
[18] tried back-to-back configuration for dual-frequency trans-
ducers to avoid the dimension limit issue and the image co-
registration problem during the catheter pulling back process,
but greater effort is needed to ensure the parallelism of two
transducers.
The three-matching-layer design of ultrasonic transducers is
beneficial for widening the bandwidth [19], which is helpful
for improving the spatial resolution. However, there is usually
a significant ring-down in the pulse echo signal and notches
 MA AND CAO: SINGLE CRYSTAL HIGH-FREQUENCY IVUS TRANSDUCER 811

TABLE I
P ROPERTIES OF T RANSDUCER M ATERIALS

Fig. 1. Schematic of the three-matching-layer transducer.

in the power spectrum, when the matching layers are not

properly designed. Manh et al. [19] tried a three-matching-
layer design, wherein they fabricated a silicone matching layer
by micromachining technology with the acoustic impedance
of the first matching layer close to 20 MRayl. Such an
acoustic impedance is inappropriate, resulting in a long 300-ns
pulse length, much longer than that of the two-matching-
layer design, although the bandwidth reached over 90%. So,
the bandwidth alone is not a good measure of the axial
resolution; we must also check the pulse length in the time
domain. In addition, the inappropriate design may also lead
to high time-sidelobes in the impulse response, causing image
degradation [20], [21].
In this investigation, we carefully designed and made a
single-crystal, high-frequency ultrasound transducer using a Fig. 2. Pulse-echo simulation result of the three-matching-layer IVUS
transducer.
three-matching-layer methodology to improve the resolution
while maintaining the short pulse length. The single-element-
transducer is embedded in a 1.5-m-long, 3.5-Fr catheter made B. Acoustic Design and Simulation
of PE material. The pulse-echo performance is evaluated. The PiezoCAD (Sonic Concepts, Inc., Bothell, WA, USA) based
wire phantom imaging and in vitro imaging through 360◦ on the KLM equivalent circuit model and Field II modeling
rotational scanning showed much improved spatial resolution software [25] are used to simulate the acoustic waveform and
than that of a similar frequency commercial transducer without the lateral resolution, respectively. The acoustic impedance
sacrificing the penetration depth. and the thickness of materials of the three-matching-layer
transducer are given in Table I.
II. M ATERIALS AND M ETHODS According to the pulse echo performance simulation shown
in Fig. 2 by PiezoCAD, the −6-dB bandwidth of the three-
A. Transducer Specification matching-layer transducer is close to 70% and the −20-dB
Our IVUS transducer is composed of three matching layers, pulse length is less than 50 nS, as listed in Table II.
one 0.72Pb(Mg1/3Nb2/3 )O3 − 0.28PbTiO3 (PMN-28%PT) A key factor for clinical imaging is the spatial resolution,
single-crystal (CTS Corporation, Bolingbrook, IL, USA) ele- which is determined by the transducer frequency and the effec-
ment, and a conductive backing. The schematic diagram of tive aperture [26]. Based on the Field II software pulse-echo
the transducer is shown in Fig. 1. The high piezoelectric field simulation, the lateral resolution (−6-dB beamwidth) of
charge constants of the PMN-28%PT single crystal without the three-matching-layer transducer at the depth of 3 mm is
grain boundaries are suitable for high-frequency applications about 210 μm, as illustrated in Fig. 3.
[22], [23].
The transducer has an effective aperture of
C. Transducer Fabrication and Characterization
0.4 mm × 0.5 mm, which has a proper and regular
size for balancing the catheter fitness, electrical impedance, We have taken into account the bandwidth and the pulse
and spatial resolution for IVUS application, and the FC is length in the design optimization, the optimized acoustic
designed to be 45 MHz, aiming to have an enhanced spatial impedance and the thickness of the three matching layers are
resolution [18], [24]. given in Table I.

Authorized licensed use limited to: Hong Kong University of Science and Technology. Downloaded on August 14,2020 at 15:14:21 UTC from IEEE Xplore. Restrictions apply.
 812 IEEE TRANSACTIONS ON ULTRASONICS, FERROELECTRICS, AND FREQUENCY CONTROL, VOL. 67, NO. 4, APRIL 2020
TABLE II
P ULSE -E CHO S IMULATION OF THE T HREE -M ATCHING -L AYER
T RANSDUCER
Fig. 3. Lateral resolution simulation result at 3-mm depth.
To avoid the bond-line effect on the acoustic performance
[27], the second matching and backing layers were cast and
then lapped down to the specified thickness. The detailed
fabrication process is as follows.
1) The single-crystal material with 5 mm × 5 mm aperture
is lapped down to 150 μm. For the electrodes, a 0.1-μm
Ni–Cr layer is coated on both sides followed by a gold
coating layer of 0.2 μm in thickness.
2) The backing material E-Solder 3022 conductive epoxy
(VonRoll Isola, New Haven, CT, USA) is cast on one
side of the single crystal. The backing is cured overnight
at 40 ◦ C in a forced convection oven. Then, the backing
material is lapped down to 500 μm thickness.
3) The single crystal is lapped down to the specified thick-
ness of 40 μm and the exposed single crystal surface
is plated with 0.1-μm Ni–Cr and 0.2-μm gold layers as
the electrode.
4) After this, the first matching layer of aluminum alloy
is bonded on the plated single crystal surface using
301-2 epoxy (Epoxy Technology, Inc., Billerica, MA,
USA) and cured overnight in a forced convection oven
at 40 ◦ C. Then, the matching layer is lapped down to
the specified thickness of 30 μm. The thickness of the
current stack is 570 μm and is recorded as the reference
for the second matching layer processing.
5) The second matching layer is composed of 2–3-μm
silver particles (Sigma-Aldrich, Inc., St. Louis, MO,
USA) mixed with Insulcast 502 & Insulcure 9 epoxy
(American Safety Technologies, Roseland, NJ, USA)
and cured overnight in a forced convection oven at
40 ◦ C. The ratio of the epoxy to the silver particle
is 1:2.4 by weight and the thickness of the second
Authorized licensed use limited to: Hong Kong University of Science and Technology. Downloaded on August 14,2020 at 15:14:21 UTC from IEEE Xplore. Restrictions apply.
Fig. 4. IVUS transducer stack profile (before Parylene C coating).
Fig. 5. Electrical impedance plot of the IVUS transducer stack.
matching layer is lapped down to 11 μm. To minimize
the oxidation of silver particles, 0.1-μm Ni–Cr and
0.2-μm gold layers are sputtered on the surface of
the second matching layer.
6) Before connecting to the coaxial cable, the stack is
diced orthogonally using a Disco Dicing Saw DAD322
(Disco Corporation, Tokyo, Japan) into multiple indi-
vidual pillars having an aperture of 0.4 mm × 0.5 mm.
Each layer is clearly shown in Fig. 4. The transducer
electrical impedance without cable is measured using
an Agilent 4294A Impedance Analyzer (Keysight Tech-
nologies, Santa Rosa, CA, USA). Because of multiple
matching layers, the impedance magnitude shows some
wiggling around 30 , instead of a clear resonance and
antiresonance, as shown in Fig. 5.
7) Finally, the IVUS transducer is connected to a
46-AWG coaxial cable (TE Connectivity, Schaffhausen,
Switzerland) of 50- characteristic impedance. Then,
a 12-μm Parylene C layer is coated on the transducer
surface as the third matching layer and also serves as
the protective layer. The whole transducer is embedded
in a 1.5-m-long, 3.5-Fr catheter shown in Fig. 6.
D. Setup and Testing Method
1) Pulse Echo Setup: The acoustic waveform is obtained by
the pulse-echo test method. A stainless steel block is used as
 MA AND CAO: SINGLE CRYSTAL HIGH-FREQUENCY IVUS TRANSDUCER 813

Fig. 6. Photograph of the three-matching-layer IVUS catheter.

the target and the transducer is excited by a spike pulse from

a DPR500 pulser/receiver (JSR SONIX Ultrasonics, Pittsford,
NY, USA) with 2000-Hz pulse repetition frequency, 8-μJ
energy per pulse, 10-dB gain, 5–500-MHz bandpass filters,
and 50- damping. The reflected waveform is received and
digitized by a 500-MHz oscilloscope (MDO3052, Tektronix Fig. 7. Two-wire resolution phantom test configuration of (a) axial
Corporation, Beaverton, OR, USA) with 2.5 Gs/s sampling rate resolution and (b) lateral resolution.
and 50- coupling. The echo signal data are transferred from
the oscilloscope to a computer and the waveform displayed
in the time and frequency domain plots were generated by a
customized MATLAB code. The loop sensitivity (LS), the FC,
and the fractional bandwidth (BW) are defined as follows:
 
Vreceive
LS = 20 × log10 (1)
Vtransmit
Fl + Fu
FC = (2)
2
Fu − Fl
BW = × 100% (3)
FC
where Vreceive is the peak-to-peak voltage in V of the echo
from the target and Vtransmit is the DPR500 pulser out-
put voltage. Fl and Fu are defined as lower and upper Fig. 8. Pulse-echo performance of the three-matching-layer IVUS
−6-dB frequencies, respectively, at which the magnitude of transducer.
the frequency spectrum is 50% (−6 dB) of the maximum.
2) Spatial Resolution and Imaging Test Setup: The spatial
resolution was measured and in vitro imaging was obtained by
connecting the three-matching-layer IVUS transducer to a cus-
tomized IVUS imaging platform. A DPR500 pulser/receiver
was used to drive the transducer for pulse generation and
echo detection. RF data were digitized using a 12-bit data
acquisition board (Gage Applied Technologies, Lockport, IL,
USA) with a sampling rate of 200 MHz. The 500 A-lines
were acquired during each revolution and the scanning pro-
cedure was controlled using a customized LabVIEW program
(National Instruments, Austin, TX, USA). The RF data were
saved and post-processed for the image display with a 50-dB
dynamic range.
Two tungsten wires of 10-μm diameter are used as the phan- Fig. 9. Pulse-echo performance of the commercial IVUS transducer.
tom for both axial and lateral spatial resolution measurements.
The phantom testing configuration is shown in Fig. 7. k33 mode [28], but based on the pulse-echo performance
comparison between results in Figs. 8 and 9, the measured
III. E XPERIMENTAL R ESULTS waveform of the three-matching-layer, single-crystal IVUS
Transducer performance evaluation results are given below, transducer has much broader bandwidth and comparable LS
including the acoustic waveform test, spatial resolution mea- compared with the commercial IVUS transducer.
surement, and in vitro artery imaging.
B. Spatial Resolution Measurement
A. Pulse-Echo Test Due to the broad bandwidth and short pulse length, the axial
The measured results are given in Figs. 8 and 9 and resolution of the three-matching-layer transducer reached
in Table III. Since the piezoelectric single crystal in the 40 μm. As can be seen in Fig. 10, two tungsten wires with the
transducer works in the kt plate mode instead of the k33 separation of 40 μm marked by the yellow circle can easily
rod mode, the bandwidth is not as high as that of the be distinguished. While the commercial IVUS transducer

Authorized licensed use limited to: Hong Kong University of Science and Technology. Downloaded on August 14,2020 at 15:14:21 UTC from IEEE Xplore. Restrictions apply.
 814 IEEE TRANSACTIONS ON ULTRASONICS, FERROELECTRICS, AND FREQUENCY CONTROL, VOL. 67, NO. 4, APRIL 2020
TABLE III
C OMPARISON OF M EASURED P ULSE -E CHO P ERFORMANCE
OF THE T HREE -M ATCHING -L AYER T RANSDUCER AND
THE C OMMERCIAL T RANSDUCER
Fig. 10. Axial resolution measurement of the three-matching-layer
transducer.
Fig. 11. Axial resolution measurement of the commercial transducer.
(Boston Scientific SR Pro2) cannot distinguish such 40-μm
separation, as shown in Fig. 11.
According to the Field II simulation results, the lateral reso-
lution of the three-matching-layer transducer is about 210 μm.
So, two tungsten wires with 220-μm spacing are positioned
at 3 mm from the transducer for lateral resolution verification.
Based on the comparison between Figs. 12 and 13, one can
see that the three-matching-layer transducer can distinguish
the two wires, while the commercial product cannot.
In order to get more accurate values, the point spread
functions (PSFs) of a wire phantom are used to calculate the
axial and lateral resolutions. A 10-μm-diameter tungsten wire
Authorized licensed use limited to: Hong Kong University of Science and Technology. Downloaded on August 14,2020 at 15:14:21 UTC from IEEE Xplore. Restrictions apply.
Fig. 12. Lateral resolution measurement of the three-matching-layer
transducer.
Fig. 13. Lateral resolution measurement of the commercial transducer.
Fig. 14. Resolution comparison between the three-matching-layer
transducer and the commercial transducer. (a) Axial and (b) lateral
envelops.
target is linearly scanned to determine the axial and lateral
resolutions with a distance of 3 mm from the transducer
surface. The comparison of PSF echo signals between the
three-matching-layer transducer and the commercial one is
displayed in Fig. 14(a) and (b).
From Table IV, the axial resolution of the
three-matching-layer transducer is only half of that of the
commercial transducer, which is a noticeable improvement for
IVUS application, and the lateral resolution is also improved
slightly, which verified the Field II simulation results.
Fig. 15(a) shows a tissue-mimic gelatin phantom used for
the penetration depth comparison test. The experimental con-
 MA AND CAO: SINGLE CRYSTAL HIGH-FREQUENCY IVUS TRANSDUCER
TABLE IV
S PATIAL R ESOLUTION C OMPARISON
Fig. 15. (a) Tissue-mimic gelatin phantom. (b) B-mode image of the
three-matching-layer transducer. (c) B-mode image of the commercial
transducer.
Fig. 16. Healthy porcine heart for the i n vitro imaging test.
figuration is the same as the two-wire spatial resolution setup.
As shown in Fig. 15(b) and (c), both transducers can detect
the signals reflected at the depth of 5 mm, which is a standard
depth of penetration requirement in IVUS applications [29].
So, the three-matching-layer IVUS transducer has a compara-
ble penetration but much superior spatial resolutions compared
with the commercial transducer.
C. In vitro Imaging Evaluation
The transducer-embedded catheter is inserted into the coro-
nary artery of a healthy porcine heart (see Fig. 16) for the in
Authorized licensed use limited to: Hong Kong University of Science and Technology. Downloaded on August 14,2020 at 15:14:21 UTC from IEEE Xplore. Restrictions apply.
815
Fig. 17. Coronary artery imaging using the three-matching-layer
transducer.
Fig. 18. Coronary artery imaging using the commercial IVUS transducer
(Boston Scientific SR Pro2).
vitro imaging performance evaluation of the three-matching-
layer IVUS transducer. A commercial IVUS catheter (Boston
Scientific SR Pro2) is used for the imaging quality comparison.
From the comparison between Figs. 17 and 18, one can
see that the spatial resolution of the three-matching-layer
IVUS transducer is better than that of the commercial IVUS
transducer, while the penetration depth is comparable.
IV. C ONCLUSION
A three-matching-layer transducer has been designed and
successfully fabricated using the PMN-28%PT single crystal.
Pulse-echo waveform measurement, spatial resolution mea-
surement, and in vitro imaging were performed using this
transducer. Compared with the standard commercial product,
our three-matching-layer transducer has 38% broader band-
width, which can produce better resolution for clinical IVUS
applications. By connecting to a customized IVUS imaging
platform, the axial and lateral resolutions of our transducer
have reached 41.6 and 214.7 μm, respectively. Such high res-
olution is potentially capable of detecting a thin fibrous cap of
less than 60-μm thickness. It is known that the thickness of the
fibrous cap is one of the main parameters for doctors to distin-
guish vulnerable plaques [30]. Clinically, 60 μm is considered
 816 IEEE TRANSACTIONS ON ULTRASONICS, FERROELECTRICS, AND FREQUENCY CONTROL, VOL. 67, NO. 4, APRIL 2020
a critical thickness for doctors to determine if the patient
should be treated by stenting without causing plaque rupture.
The advantage of our transducer is testified through the
in vitro imaging of a porcine heart coronary artery; a much
sharper image was produced compared with the one produced
by a standard commercial IUVS transducer (Boston Scientific
SR Pro2). All test results showed that our three-matching-layer
transducer is a promising candidate for replacing the current
commercial IVUS transducer for better imaging quality, and its
axial resolution has reached the level that could help doctors
to distinguish vulnerable atherosclerotic plaques.
R EFERENCES
[1] E. J. Benjamin et al., “Heart disease and stroke statistics-2018 update:
A report from the american heart association,” Circulation, vol. 137,
no. 12, p. e67, Jan. 2018.
[2] M. J. Kern et al., “Physiological assessment of coronary artery disease in
the cardiac catheter-ization laboratory: A scientific statement from the
american heart associationcommittee on diagnostic and interventional
cardiac catheterization,” Councilon Clin. Cardiol., Circulat., vol. 114,
pp. 1321–1341, 2006.
[3] G. S. Mintz et al., “American College of Cardiology clinical expert
consensus document on standards for acquisition, measurement and
reporting of intravascular ultrasound studies (ivus): A report of the
american college of cardiology task force on clinical expert consensus
documents,” J. Amer. College Cardiol., vol. 37, no. 5, pp. 1478–1492,
2001.
[4] H. M. Garcia-Garcia, M. A. Costa, and P. W. Serruys, “Imaging
of coronary atheroscle-rosis: Intravascular ultrasound,” Eur. Heart J.,
vol. 31, pp. 2456–2469, 2010.
[5] K. Jansen, G. van Soest, and A. F. van der Steen, “Intravascular
photoacoustic imaging: A new tool for vulnerable plaque identification,”
Ultrasound Med. Biol., pp. 1037–1048, 2014.
[6] M. R. Elliott and A. J. Thrush, “Measurement of resolution in intravas-
cular ultrasound images,” Physiolog. Meas., vol. 17, no. 4, pp. 259–265,
Nov. 1996.
[7] B. ME et al., “Assessing atherosclerotic plaque morphology: Compari-
son of optical coherence tomography and high frequency intravascular
ultrasound,” Heart, vol. 77, pp. 397–403, 1997.
[8] J. R. Yuan et al., “Microfabrication of piezoelectric composite ultrasound
transducers (PC-MUT),” in Proc. IEEE Ultrason. Symp., Oct. 2006,
pp. 922–925.
[9] X. Li et al., “80-MHz Intravascular Ultrasound Transducer Using PMN-
PT Free-standing Film,” IEEE Trans. Ultrason., Ferroelectr., Freq.
Control, vol. 58, no. 11, pp. 2281–2288, Nov. 2011.
[10] J. Li et al., “Integrated IVUS-OCT for real-time imaging of coronary
atherosclerosis,” JACC Cardiovascular Imag., vol. 7, pp. 101–103, 2014.
[11] J. Lee, J.-Y. Moon, and J. H. Chang, “A 35MHz/105MHz dual-element
focused transducer for intravascular ultrasound tissue imaging using the
third harmonic,” Sensors, vol. 18, 2018, Art. no. 2290.
[12] X. Li, J. Yin, C. Hu, Q. Zhou, K. K. Shung, and Z. Chen, “High-
resolution Coregistered intravascular imaging with integrated ultrasound
and optical coherence tomography probe,” Appl. Phys. Lett., vol. 97,
Sep. 2010, Art. no. 133702.
[13] J. Yin et al., “Novel combined miniature optical coherence tomography
ultrasound probe for in vivo intravascular imaging,” J. Biomed. Opt.,
vol. 16, 2011, Art. no. 060505.
[14] C. V. Bourantas et al., “Hybrid intravascular imaging: Current applica-
tions and prospective potential in the study of coronary atherosclerosis,”
J. Amer. College Cardiol., vol. 61, pp. 1369–1378, 2013.
[15] D. Maresca, S. Adams, B. Maresca, and A. F. van der Steen, “Map-
ping intravascular ultrasound controversies in interventional cardiology
practice,” PLoS ONE, vol. 9, no. 5, 2014, Art. no. e97215.
[16] J. Yuan, S. Rhee, and X. Jiang, “60 MHz PMN-PT based 1-3 composite
transducer for IVUS imaging,” in Proc. IEEE Int. Ultrason. Symp.,
Nov. 2008, pp. 682–685.
[17] J. Ma et al., “Design factors of intravascular dual frequency transducers
for super-harmonic contrast imaging and acoustic angiography,” Phys.
Med. Biol., vol. 60, no. 9, pp. 3441–3457, 2015.
[18] T. Ma, M. Yu, Z. Chen, C. Fei, K. K. Shung, and Q. Zhou, “Multi-
frequency intravascular ultrasound (IVUS) imaging,” IEEE Trans. Ultra-
son., Ferroelectr., Freq. Control, vol. 62, no. 1, pp. 97–107, Jan. 2015.
Authorized licensed use limited to: Hong Kong University of Science and Technology. Downloaded on August 14,2020 at 15:14:21 UTC from IEEE Xplore. Restrictions apply.
[19] T. Manh, A.-T. T. Nguyen, F. T. Johansen, and L. Hoff, “Microfab-
rication of stacks of acoustic matching layers for 15 MHz ultrasonic
transducers,” Ultrasonics, vol. 54, pp. 614–620, 2014.
[20] T. L. Szabo, Diagnostic Ultrasound Imaging: Inside Out. 1st ed.
Amsterdam, The Netherlands: Elsevier, 2004.
[21] K. K. Andersen, M. Frijlink, and L. Hoff, “Numerical optimization of
ultrasound transducers by the linearity of the phase spectrum,” in Proc.
IEEE Int. Ultrason. Symp. (IUS), Sep. 2017, pp. 1–4.
[22] J. Tian, K. Meneou, B. Stone, and P. Han, “Piezoelectric crystal
composite for high frequency ultrasound application,” in Proc. IEEE
Int. Ultrason. Symp., Oct. 2010, pp. 65–67.
[23] X. Li, T. Ma, J. Tian, P. Han, Q. Zhou, and K. K. Shung, “Microma-
chined PIN-PMN-PT crystal composite transducer for high-frequency
intravascular ultrasound (IVUS) imaging,” IEEE Trans. Ultrason., Fer-
roelectr., Freq. Control, vol. 61, no. 7, pp. 1171–1178, Jul. 2014.
[24] B. Zhu, W. Wei, Y. Li, X. Yang, Q. Zhou, and K. K. Shung, “KNN-
based Single Crystal High Frequency Transducer for Intravascular
Photoacoustic Imaging,” in Proc. IEEE Int. Ultrason. Symp., Sep. 2017,
pp. 1–4.
[25] J. A. Jensen, “FIELD: A program for simulating ultrasound systems,” in
Proc. IEEE 10th Nordic-Baltic Conf. Biomed. Imag., vol. 4, Mar. 1996,
pp. 351–353.
[26] R. E. McKeighen, “Design guidelines for medical ultrasonic arrays,”
Proc. SPIE, vol. 3341, May 1998.
[27] H. Li, Y. C. Li, D. Zhou, J. Peng, H. S. Luo, and J. Y. Dai, “Application
of PMNPT single crystal in a 3.2 MHz phased-array ultrasonic medical
imaging transducer,” in Proc. 16th IEEE Int. Symp. Appl. Ferroelectr.,
May 2007, pp. 572–574.
[28] M. Kim, J. Kim, and W. Cao, “Experimental technique for characterizing
arbitrary aspect ratio piezoelectric resonators,” Appl. Phys. Lett., vol. 89,
2006, Art. no. 162910.
[29] J. Waggoner and D. M. Feldman, “How Do OCT and IVUS differ?
A comparison and assessment of these modern imaging modalities,”
Cardiac Interventions Today, pp. 46–52, May/Jun. 2011.
[30] K. Fujii, H. Hao, M. Ohyanagi, and T. Masuyama, “Intracoro-
nary imaging for detecting vulnerable plaque,” Circulat. J., vol. 77,
pp. 588–595, Mar. 2013.
Xiaowen Ma received the B.S. degree in bio-
medical engineering from Shanghai Jiaotong
University, Shanghai, China, in 2008, and the
M.S. degree in electronic and communication
engineering from Fudan University, Shanghai, in
2012. He is currently pursuing the Ph.D. degree
in biomedical engineering with the School of
Life Science and Technology, Harbin Institute of
Technology, Harbin, China.
His current research interests include the
design and fabrication of high frequency medical
ultrasonic transducers and the application of new piezoelectric materials
in ultrasonic transducers.
Wenwu Cao received the B.S. degree in physics
from Jilin University, Changchun, China, in
1982, and the Ph.D. degree in condensed mat-
ter physics from Pennsylvania State University,
State College, PA, USA, in 1987.
He is currently a Professor of mathematics and
materials science, a joint appointment between
the Department of Mathematics and the Materi-
als Research Institute, Pennsylvania State Uni-
versity. He has published more than 630 articles.
He has coauthored five books with over 17000
citations. He has conducted both theoretical and experimental research
in condensed matter physics, materials science, medical ultrasound,
and device engineering, including theories on proper- and improper-
ferroelastic phase transitions, static and dynamic properties of domains
and domain walls in ferroelectric and ferroelastic materials, measure-
ments on second and third order elastic constants, linear and nonlinear
dielectric constants and piezoelectric constants in single crystals and
ceramics, static and dynamic behavior of piezoelectric ceramic-polymer
composites, simulation design of piezoelectric sensors, transducers and
actuators for underwater acoustics and medical ultrasonic imaging,
and ultrasonic NDE. In recent years, his research has extended to
sonodynamic therapy to treat cancers, an emerging new noninvasive
cancer treatment modality.