See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/315174605

Simulation Study of an Ultrasound Retinal Prosthesis With a Novel Contact-

Lens Array for Noninvasive Retinal Stimulation

Article  in  IEEE transactions on neural systems and rehabilitation engineering: a publication of the IEEE Engineering in Medicine and Biology Society · March 2017
DOI: 10.1109/TNSRE.2017.2682923

CITATIONS READS

10 1,632

12 authors, including:

Guofeng Li Hongyang Jiang

Guangdong Medical University Northeastern University (Shenyang, China)
21 PUBLICATIONS   295 CITATIONS    15 PUBLICATIONS   470 CITATIONS   

SEE PROFILE SEE PROFILE

Congzhi Wang Feiyan Cai

Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzh… Chinese Academy of Sciences
54 PUBLICATIONS   797 CITATIONS    89 PUBLICATIONS   1,791 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Genomic sequencing View project

Image-Guided Lung Biopsy Surgery View project

All content following this page was uploaded by Weibao Qiu on 06 October 2017.

The user has requested enhancement of the downloaded file.

 IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, VOL. 25, NO. 9, SEPTEMBER 2017
Simulation Study of an Ultrasound Retinal
Prosthesis With a Novel Contact-Lens
Array for Noninvasive Retinal
Stimulation
Mengdi Gao, Yanyan Yu, Huixia Zhao, Guofeng Li, Hongyang Jiang, Congzhi Wang, Feiyan Cai,
Leanne Lai-Hang Chan, Bernard Chiu, Wei Qian, Weibao Qiu, and Hairong Zheng
Abstract — Millions of people around the world suffer from
varying degrees of vision loss (including complete blind-
ness) because of retinal degenerative diseases. Artificial
retinal prosthesis, which is usually based on electrical neu-
rostimulation, is the most advanced technology for different
types of retinal degeneration. However, this technology
involves placing a device into the eyeball, and such a highly
invasive procedure is inevitably highly risk and expensive.
Ultrasound has been demonstrated to be a promising tech-
nology for noninvasive neurostimulation, making it possible
to stimulate the retina and induce action potentials similar to
those elicited by light stimulation. However, the technology
of ultrasound retinal stimulation still requires considerable
developments before it could be applied clinically. This
paper proposes a novel contact-lens array transducer for
use in an ultrasound retinal prosthesis (USRP). The trans-
Manuscript received July 18, 2016; revised October 31, 2016;
accepted March 6, 2017. Date of publication March 15, 2017; date
of current version September 2, 2017. This work was supported
in part by the National Science Foundation Grants of China under
Grant 61302038, Grant 11325420, Grant 81527901, Grant 61571431,
Grant 11272329, Grant 11574342, and Grant 11534013, in part by the
National Basic Research Program of China under Grant 2015CB755500,
in part by Research Project of CAS under Grant YZ01507 and
Grant QYZDB-SSW-JSC018, in part by the Guangdong Innovative
and Entrepreneurial Research Team Program under Grant 2013S046,
in part by the Natural Science Foundation of Guangdong Province
under Grant 2015A030306018, Grant 2014A030313686, and
Grant 2014A030312006, in part by the Shenzhen Peacock Plan
under Grant 20130409162728468, in part by the Shenzhen International
Collaboration under Grant GJHZ20140417113430615, and in part by the
Foundation Grants of Shenzhen under Grant JCYJ20140610151856707.
(Mengdi Gao and Yanyan Yu contributed equally to this work.)
(Mengdi Gao and Yanyan Yu are co-first authors.) (Corresponding
authors: Weibao Qiu; Hairong Zheng.)
M. Gao is with the Institute of Biomedical and Health Engineering,
Shenzhen Institutes of Advanced Technology, Chinese Academy
of Sciences, Shenzhen 518055, China, and also with the
Sino-Dutch Biomedical and Information Engineering School,
Northeastern University, Shenyang 110169, China (e-mail:
hr.zheng@siat.ac.cn).
Y. Yu, L.-H. Chan, and B. Chiu are with the Department of Electronic
Engineering, City University of Hong Kong, Hong Kong.
H. Zhao, G. Li, C. Wang, F. Cai, W. Qiu, and H. Zheng
are with the Institute of Biomedical and Health Engineering,
Shenzhen Institutes of Advanced Technology, Chinese Academy
of Sciences, Shenzhen 518055, China (e-mail: wb.qiu@siat.ac.cn;
hr.zheng@siat.ac.cn).
H. Jiang and W. Qian are with the Sino-Dutch Biomedical and Infor-
mation Engineering School, Northeastern University, Shenyang 110169,
China.
Digital Object Identifier 10.1109/TNSRE.2017.2682923
1534-4320 © 2017 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.
1605
ducer was designed in the shape of a contact lens so as
to facilitate acoustic coupling with the eye liquid. The key
parameters of the ultrasound transducer were simulated,
and results are presented that indicate the achievement of
2-D pattern generation and that the proposed contact-lens
array is suitable for multiple-focus neurostimulation, and
can be used in a USRP.
Index Terms — Ultrasound neurostimulation, noninva-
sive stimulation, ultrasound retinal prosthesis (USRP),
contact-lens shape array, multiple-focus neurostimulation.
I. I NTRODUCTION
R ETINITIS pigmentosa and age-related macular
degeneration are two common outer retinal degenerative
diseases that are leading causes of blindness [1], [2]. These
diseases cause deterioration of photoreceptors in the central
area of the retina, resulting in irreversible loss of vision [3].
Neurostimulation can be used to overcome such vision loss
via an interface connected to an external sensor [4]. Many
neural cells in the inner retina remain histologically intact in
individuals with blinding diseases, and so the visual pathway
can be restored by activating these remaining alive cells [5].
A retinal prosthesis has been developed that uses an electrical
stimulus to activate the remaining retinal nerves in blind
patients and thereby elicit the sensation of vision [6], [7].
In this device, images captured by a camera are wirelessly
transmitted to an electronic device implanted in the eyeball.
The implanted device has the ability to stimulate the neurons
via several tens of microelectrodes that are attached to the
retinal tissue, producing the perception of sight [8]. However,
one difficulty encountered when fabricating such a device is
the need for all of the materials to be biocompatible and for
the device to be sufficiently small physically to be implanted
into the eyeball. Moreover, implanting such a device into
the eyeball is associated with a high surgical risk [9], [10].
Therefore, a noninvasive method for neurostimulation would
provide a great leap in the development of retinal prostheses
for the functional restoration of sight.
Recent investigations have shown that ultrasound can be
used to stimulate neural activity noninvasively [11]–[14].
Transmitting ultrasonic waves into neural tissue can pro-
vide the noninvasive and remote stimulation of neurons by
1606 IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, VOL. 25, NO. 9, SEPTEMBER 2017

activating their ion channels [15], [16]. Ultrasound-mediated

changes in neuronal activity are sufficient to trigger synaptic
transmission in brain circuits. Ultrasound has already been
demonstrated to stimulate neurons in the retina, which could
herald the development of an ultrasound retinal prosthe-
sis (USRP) [17], [18]. Action potentials can be generated when
ultrasound stimulation is applied to the retina. The recording
of visual evoked potentials in small animals demonstrates that
the retina responds to ultrasound stimulation [17]. Interneurons
beyond the photoreceptors were found to be activated by
ultrasound stimulation, which further suggests the potential of
ultrasound in restoring vision lost due to retinal degenerative
diseases [18].
Single-element focused ultrasound transducers are currently
normally used for ultrasound neurostimulation [12]–[17],
[20], [21]. However, the stimulation point is mechanically
fixed, thereby precluding the simulation of multiple points
simultaneously. Array-based transducers can be used to stim-
ulate multiple sites using transmission beamformer technol-
ogy [22], [23]. A multielement phased array was proposed
based on design considerations of a USRP [18]. However,
the transducer had a plane shape and was large (2 × 4 cm2 ),
making it difficult to use in humans. In addition, ultrasonic
coupling would be a critical issue when applying the trans-
ducer to the human eye. Fig. 1. (a) Proposed scheme for a USRP with a contact-lens array
transducer, and the total solution for the wearable device compris-
This study investigated a novel ultrasound transducer ing an antenna, transducer, and integrated circuit (IC). (b) Layout
scheme designed specifically for USRP application. The shape of 256 elements in the concave transducer. (c) The ultrasound stimulation
and dimensions of the ultrasound transducer are similar to sequence. PRF, pulse repetition frequency; SD, stimulation duration; SP,
stimulation period; BC, base curvature radius.
those of a contact lens, being a two-dimensional (2D) phased
array with a concave shape that attaches naturally to the
cornea. These characteristics result in the device sticking to
the cornea, allowing the eye liquid to provide natural acoustic
coupling between the transducer and the cornea, and thereby
avoiding the need for external coupling. Realistic parameters
of the proposed contact-lens array were simulated in this study
to check its suitability in multiple-focus neurostimulation for
a USRP. The simulation of a complex 2D stimulation pattern
is also demonstrated.

II. M ETHODS
Fig. 1(a) illustrates the proposed scheme for the USRP
based on a concave array transducer. The array can stimulate
multiple points in the retina by using transmission beamformer
technology to generate diverse patterns of excitation. The
transducer has a contact-lens shape that makes it easy to
attach to the external surface of the eyeball. The proposed
device includes an antenna to receive an external camera
signal, an integrated circuit to process that signal and drive
a transducer, and a transducer array to provide the actual
ultrasound stimulation. This transducer design is novel in that
it is of wearable size and utilizes the natural eye liquid for
acoustic coupling. The concave transducer generates multiple
focal points at a stimulation depth equaling the diameter of
the eyeball (∼24 mm) so as to provide neural stimulation on
the retina.
A simulation study was conducted to demonstrate the feasi-
bility of the novel scheme proposed herein. 2D patterns cannot
be produced on the retina using a traditional single ultrasonic
Fig. 2. Flow chart for calculating the excitation waveform and the
ultrasound field in multiple-focus stimulation.
beam since only one focal beam could be generated. A projec-
tion algorithm was employed to generate the 2D stimulation
pattern shown in Fig. 1(b) with an array transducer. Fig. 2
shows a flow chart of the overall mechanism of generating
multiple focal points.
Section A below introduces basic aspects of the stim-
ulation procedure, including the concave transducer and
the stimulation strategy. Section B describes the Random-
Superposition (RS) algorithm and the weighted Gerchberg-
Saxton (GSW) algorithm used to determine the excitation
GAO et al.: SIMULATION STUDY OF A USRP WITH A NOVEL CONTACT-LENS ARRAY FOR NONINVASIVE RETINAL STIMULATION 1607

TABLE I
T RANSDUCER PARAMETERS FOR THE USRP A PPLICATION

waveform applied to each array element. The approach for

calculating the distribution of the ultrasound field is described
in Section C.
Fig. 3. Schematic diagram for calculating the ultrasound field produced
by a single-element concave transducer.
A. Configuration of the Array Transducer
The transducer is designed to stimulate the retinal neurons Summing the contribution of N elements in the concave
noninvasively. Focused ultrasonic beams were generated by phased array yields:
the transducer and transmitted to the retina to stimulate the
N 
neurons. The transducer has a concave shape and attaches j ωρ0 u n e j ωt 1 − j krn
naturally to the cornea. The anterior-to-posterior diameter of a p(Q) = e Rn d Rdφ (2)
2π rn
n=1 sn
typical adult eyeball is 24 mm, and so the focal length of the
ultrasound stimulation was adjusted to 24 mm. The radius of where sn is the area of the nt h element, rn is the distance
curvature of the concave array transducer is 8.7 mm and the between the center of the n t h element and target point Q in
chord length is 14 mm. The elements are distributed densely the ultrasound field, and Rn is the distance from origin O to
throughout the transducer, which maximizes the space utiliza- the center of the n t h element.
tion because the transmitted ultrasound energy is proportional To obtain a 2D patterned stimulation, multiple focal points
to the size of the transducer. are set in the area of interest in advance. The excitation
The transducer consists of 256 elements of equal size, waveform of each array element for generating the desired
as shown in Fig. 1(b). The elements are distributed in nine multiple-focus field pattern is then calculated using the
circles with outer radii of 0.9, 1.8, 2.6, 3.4, 4.2, 5.0, 5.7, RS and GSW algorithms [18].
6.4, and 7 mm; the number elements within these concentric Suppose that the excitation waveform of elements forming
circles are 3, 10, 16, 22, 29, 35, 41, 47, and 53 respectively. M multiple focal points is known. Equation 2 was applied to
The distance between adjacent elements is about 0.84 mm. calculate the sound pressure at the m t h focal point:
Fig. 1(c) shows the stimulation waveform according to the 
published literature for ultrasound neurostimulation [21]. The N
j ωρ0 u n e j ωt 1 − j krm,n
pm = e dSn (3)
waveform is a pulsed type in order to minimize the time- 2π rm,n
n=1 sn
averaged ultrasound intensity. The transducer parameters are
listed in Table I. The RS algorithm is employed to calculate the initial
excitation waveform. In this calculation, the amplitude of the
B. Determining the Excitation Waveform of Each Element pulse generated by the n th element was set to unity (|u n | ≡ 1),
which can be taken as the phased stimulation pattern. Thus,
Spherical coordinates are used here to simplify the expres- 0 |u n |
the constant item j ωρ2π Sn is discarded, and Equation 3 can
sion for the ultrasound field generated by the concave trans-
be simplified to:
ducer. The acoustic pressure from source point P to target
point Q shown in Fig. 3 can be obtained through transforma- 
N

tion of the Rayleigh-Sommerfeld diffraction integrals [24]: pm = r−1

m,n e
j (ϕn −k·rm,n +ωt )
(4)
  n=1
j ωρ0 ue j ωt b 2π 1 − j kr  Re( pm )
p(Q) = e R1 d R1 dφ (1) The derivation ∂ = 0 is used to acquire the
2π 0 0 r ∂ϕn
M
where R1 d R1 dφ is the integral area, ρ0 is the density of the maximal summation of the real part of the complex amplitude,
medium, k is the wave number, e j ωt is the time-dependence and the initial phases of elements can be obtained:
item ( j 2 = −1), ω is the frequency of the acoustic wave, r is 
M
the distance between points p and Q, and u is the stimulation UnS R = r−1
m,n e
j k·dm,n
(5)
provided to the element. m=1
1608 IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, VOL. 25, NO. 9, SEPTEMBER 2017

Fig. 5. (a) Relationship between the ultrasound frequency and the lateral
resolution. (b) Lateral resolution of a single focus for an ultrasound center
frequency of 2.5 MHz. FWHM, full width at half maximum.

and φn is the azimuth angle of its orthogonal projection on a

reference plane that passes through the origin and is orthogonal
to the zenith, measured from a fixed reference direction on
that plane. The dw and dh parameters indicate the width and
height, respectively, of a single element.
Fig. 4. Schematic diagram for calculating the ultrasound field produced
by an array concave transducer. The second coordinate system has the center of the
n t h array element as its origin. The element is divided into
t × t subelements that can be handled as point sources. The
The phase excitation determined by the simple one-step radius of the ultrasound field for each point source should be
RS algorithm has poor uniformity. The uniformity of the power far lower than the sound wavelength, and d x ·d y is the surface
distributedto each focal point is quantified by Equation 6. area of an infinitesimal subelement:
L
pm = S · l=1 pm,l is the estimate of the power delivered to
dx · d y = R 2 · sin θ · dθ · dφ (10)
the m t h focal point computed by summing the L pixels that
cover that focal point, where S is the area of pixels in the where R is the radius of the sphere. The pressure field of the
stimulated ultrasound field map: concave spherical transducer can then be obtained by summing
pmax − pmin the pressure components of all the infinitesimal subelements.
u= (6) The acoustic pressure at observation point Q in the field can
pmax + pmin
be calculated as:
where pmax = max { pm } and pmin = min { pm }.  θ φ
The GSW algorithm is applied to further improve the N
j ωρ0 u n e j ωt + 2 + 2 1 − j krn 2
p(Q) = e R
uniformity of the complex pattern stimulation. The amplitude 2π − θ2 − φ2 rn
n=1
of the acoustic pressure field is summed for each single focus: · sin θ · dθ · dφ (11)
  N 
M  
 −1 j (ϕn −k·rm,n )  where rn is the distance between the center of the subelement
pm = rm,n e  (7)
m=1 n=1
and the observation point in the field, and u n is the excitation
of the n t h of the phased array that can be calculated using
To achieve the maximal summation, we can update the Equation 8.
excitation as:

M t −1 III. R ESULTS
t  pm 
UnG S W (t) = wm · e j k·dm,n (8)
 t −1  A. Analysis of the Resolution at Different
m=1  pm 
Ultrasonic Frequencies
where The lateral resolution of a single focal point located 24 mm
M t −1
1 pm along the Z -axis was calculated in order to evaluate the
t
wm = · 
m=1  (9)
M  t −1  quality of the focus. The lateral resolution can be quantified
 pm 
by calculating the mean full width at half maximum (FWHM)
of the focal point. Fig. 5(a) shows that the lateral resolution
C. Superposition of the Ultrasound Field of the focal point changed from 0.5 and 12.5 mm when the
Fig. 4 shows a schematic diagram of the coordinate system ultrasound frequency was changed from 0.3 to 6 MHz. The
and geometry of the transducer. The transducer is present change in the resolution of the focal point was greatest for
in a three-dimensional (3D) space whose center (x n , yn ) is frequencies below 2.5 MHz, and so 2.5 MHz was used as
described by three parameters (rn , θn , φn ) in a spherical coor- the center frequency for ultrasound neurostimulation in this
dinate system, where rn and θn represent the distance from the simulation study. Fig. 5(b) displays the normalized ultrasound
center of the n t h element to the fixed origin point and the polar pressure for 2.5 MHz ultrasound, and shows that the spatial
angle measured from a fixed zenith direction, respectively, resolution (at −6 dB) was 1.3 mm.
GAO et al.: SIMULATION STUDY OF A USRP WITH A NOVEL CONTACT-LENS ARRAY FOR NONINVASIVE RETINAL STIMULATION
Fig. 6. (a) Acoustic pressure distribution of three focal points in the
X-Y plane. (b) Distribution of areas in which the power was higher than
the threshold required to potentially activate neurons.
Fig. 7. (a) Sparse distribution of multiple sites set in advance.
(b) Distribution of simulated sparse foci in the ultrasound field. (c) Dense
distribution of multiple sites set in advance. (d) Distribution of stimulated
dense foci in the ultrasound field.
B. Construction of Multiple Focal Points
By applying the calculated waveform used to excite each
transducer element, three focal points can be generated
simultaneously and the distribution of the ultrasound field
can be obtained, as shown in Fig. 6(a). The focal points
are separated by 3 mm in the axial and lateral directions.
Considering the maximum value of the spatial peak pulse
average intensity (I S P P A ) to be 10 W/cm2 , I S P P A of the
three focal points were 9.25, 10, and 8.09 W/cm2 , and the
spatial peak temporal average intensity (I S PT A ) values were
0.77, 1.08, and 0.67 W/cm2 , respectively. These values were
obtained with a pulse repetition frequency of 1 kHz, a stimu-
lation duration of 300 ms, and a stimulation period of 3 s.
The threshold level for activating the neurons in the retina
is around 0.25 W/cm2 [17]. Assuming that the neurons would
not to be activated for ultrasound powers below that threshold,
the activation area in the experiment involving three focal
points could be updated to that shown in Fig. 6(b). The average
size of the three focal points in Fig. 6(b) is 1.3∗ 1.6 mm2 when
setting threshold level to 0.25 W/cm2 .
C. 2D Pattern Generation
In another simulation, multiple equispaced focal points were
set in front of the transducer in the pattern of the ‘T’ character
(Fig. 7(a)). The acquired ultrasound field in the X-Y plane
1609
Fig. 8. (a) 3D visualization of the ultrasound field with compensated focal
points in the X-Y plane. (b) Pressure distribution of the compensated
focal points in the X-Y plane.
is shown in Fig. 7(b), demonstrating the focal points in a
“T” pattern in the target plane.
In some cases a continuous stimulation is needed.
To achieve this we decreased the distance between the adjacent
focal points and increased the number of foci from 9 to 33 cov-
ering the same area, as shown in Fig. 7(c). The resulting
improved continuity of the focal area can be seen in Fig. 7(d).
D. Intensity Uniformity
The iteration process of the GSW algorithm was applied
to optimize the uniformity of the amplitude at different focal
points. The weighting of each focal point was updated and
a new excitation waveform was generated to ensure that
the energy was distributed evenly among the target points.
The threshold for uniformity fluctuation was set at 0.95, and
the iteration process was terminated if the uniformity of the
nine points exceeded that threshold. This process achieved a
uniform energy distribution across the different focal points,
as shown in Fig. 8
E. Influence of Depth Differences
The retina is not flat (Fig. 1(a)), and so multiple focal
points need to be positioned at different depths for this curved
structure. We therefore recalculated the stimulation pattern by
changing the focal points to A (0, −6, −15.3), B (0, 0, −12.2),
and C (0, 6, −10.7), which have the same X value and different
stimulation depths. Here the stimulation depth equals value
of the Z coordinate plus the radius of curvature (8.7 mm),
because the original point is in the geometrical center point.
The three points are shown in Fig. 9(a).
Instead of calculating the ultrasound field for the X-Y plane,
we compared the focus at different depths by analyzing the
ultrasound field in the Y -Z plane. The 3D ultrasound pressure
amplitude and 2D top view are displayed in Fig. 9(b) and (c),
respectively. For a specified stimulation depth, the resolution
of the focal point changes with the depth in the ultrasound
field. The peak pressure is highly consistent with the set
target, but it has a cigar shape in the axial direction, which
is similar to the findings of studies of ultrasound neurostim-
ulation [25], [26]. Fig. 9(d) illustrates that when the stimu-
lation depth was set at 24 mm, the resolution decreased for
depths up to 24 mm and then increased slightly at greater
depths.
1610 IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, VOL. 25, NO. 9, SEPTEMBER 2017
Fig. 9. (a) Distribution of multiple sites set in advance at dif-
ferent depths. (b) 3D visualization of the acoustic pressure in the
Y-Z plane. (c) Pressure distribution of three focal points in the Y-Z plane.
(d) Relationship between depth in the ultrasound field and the resolution
for a single focal point at a stimulation depth of 24 mm.
IV. D ISCUSSION
A retinal prosthesis provides an effective solution for restor-
ing some degree of light perception in blind patients by
activating the remaining live neurons in the inner retina. The
approval of a commercial prosthesis device by the FDA has
opened a new era of this technology [27]. However, the use
of highly invasive procedures remains the main drawback.
Ultrasound has recently been applied in human tests for non-
invasive neurostimulation [28]. Preliminary investigations of
retinal stimulation have also shown the promise of ultrasound
for use in a noninvasive retinal prosthesis [17], [18]. Previous
study [17] has shown that the power intensity of ultrasound
needs to be higher than a certain threshold (∼0.25 W/cm2 )
to generate action potentials. An ultrasound device should
generate focal points in the retina tissue at intensities higher
than this threshold. In addition, a method based on an array
ultrasound transducer is required to provide electrical stimula-
tion that will appear as an image on the retina tissue. Moreover,
the ultrasound stimulation needs to be acoustically coupled to
the retina, which would be a major difficulty for a USRP.
This paper has proposed a novel array transducer with a
contact-lens shape that facilitates acoustic coupling. A 2D
image produced using multiple focal points was simulated,
which demonstrated the feasibility of the proposed method.
The next step could be to fabricate a 2D array ultrasound
device for use in a real USRP device based on the method
proposed in this paper. MRI compatible materials should be
included for effective evaluation by MRI machine.
Similar to the retinal prosthesis device approved by the
FDA, an antenna is required to receive the data and pro-
vide power wirelessly for a USRP device (as demon-
strated in Fig. 1(a)). An application-specific integrated circuit
realizing the proposed algorithms should be designed to
process the data and perform the ultrasound neurostimulation.
The concave array transducer with 256 elements could be fab-
ricated using composite PZT materials. The distance between
adjacent elements (pitch) is set to 0.84 mm, about 1.36 λ
for suppressing the side-lobes. The power efficiency could be
optimized using 1-3 composite in a single-crystal device [29].
There is no FDA regulation for ultrasound neuro-stimulation
yet although some human tests have already been reported.
According to the published papers [17], [18], it’s safe to
stimulate the retina tissue with relatively high power ultra-
sound, although the power limit for ophthalmic ultrasound is
17 mW/cm2 [19]. Ultrasound stimulation induced heating was
only ∼0.5 Celsius degree [17]. Histological section revealed
no visible damage to the retina tissue [18].
It is difficult to generate multiple focal points simultane-
ously with the traditional transmission beamformer method.
We therefore used a method that is suitable for generating mul-
tiple focal points and thereby produce a patterned stimulation
in two dimensions. The ultrasound simulation is expected to
generate an effect similar to that obtained when using phys-
ical electrodes. The ultrasound intensity needs to be higher
than a certain threshold to produce effective neurostimulation.
Fig. 6(b) clearly shows that the intensity was sufficiently high
only in the central focal points. Although some side lobes
are evident, they should have no effect if the intensity is
lower than the stimulation threshold, and so a reasonable
2D stimulation pattern can be generated. The 2D resolution,
i.e. the average size of three focal points is 1.3∗1.6 mm2 when
setting a threshold level to 0.25 W/cm2 .
The distribution of the ultrasound focal points was found
to be better in the X-Y plane than in the Z direction. It has
a cigar shape, as shown in Fig. 9(c). The ultrasound energy
covered all of the layers of the retina and so would be able to
activate the neurons in different layers.
V. C ONCLUSION
A proposed novel contact-lens array for use in a USRP
application was simulated in this study. The intrinsic noninva-
siveness of this approach makes it a promising technique for
use in a retinal prosthesis. The simulation results show that the
device can achieve multiple focal points in two dimensions
simultaneously, which demonstrates the potential to provide
patterned stimulation in real retinal tissue. This study has
theoretically verified the feasibility of patterned stimulation
using novel algorithms in a noninvasive retinal prosthesis.
R EFERENCES
[1] E. L. Berson, “Retinitis pigmentosa. The friedenwald lecture,” Invest.
Ophthalmol. Vis. Sci., vol. 34, no. 5, pp. 1659–1676, 1993.
[2] R. Klein, B. E. Klein, S. C. Jensen, and S. M. Meuer, “The five-year
incidence and progression of age-related maculopathy: The Beaver dam
eye study,” Ophthalmology, vol. 104, no. 1, pp. 7–21, 1997.
[3] R. K. Sharma and B. Ehinger, “Management of hereditary retinal
degenerations: present status and future directions,” Survey Ophthalmol.,
vol. 43, no. 5, pp. 427–444, 1999.
[4] J. D. Weiland, W. Liu, and M. S. Humayun, “Retinal prosthesis,” Annu.
Rev. Biomed. Eng., vol. 7, pp. 361–401, Aug. 2005.
[5] J. M. Ong and L. da Cruz, “The bionic eye: A review,” Clin. Experim.
Ophthalmol., vol. 40, no. 1, pp. 6–17, 2012.
[6] 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, 2014.
GAO et al.: SIMULATION STUDY OF A USRP WITH A NOVEL CONTACT-LENS ARRAY FOR NONINVASIVE RETINAL STIMULATION
[7] J. D. Weiland and M. S. Humayun, “Retinal prosthesis,” IEEE Trans.
Biomed. Eng., vol. 61, no. 5, pp. 1412–1424, May 2014.
[8] M. S. Humayun et al., “Visual perception in a blind subject with a
chronic microelectronic retinal prosthesis,” Vis. Res., vol. 43, no. 24,
pp. 2573–2581, 2003.
[9] S. Picaud and J. A. Sahel, “Retinal prostheses: Clinical results and future
challenges,” Compet Rendus Biol., vol. 337, no. 3, pp. 214–222, 2014.
[10] J. O. Winter, S. F. Cogan, and J. F. Rizzo, “Retinal prostheses: Current
challenges and future outlook,” J. Biomater. Sci., Polymer Ed., vol. 18,
no. 8, pp. 1031–1055, 2007.
[11] S. Ibsen, A. Tong, C. Schutt, S. Esener, and S. H. Chalasani, “Sono-
genetics is a non-invasive approach to activating neurons Caenorhabditis
elegans,” Nature Commun., vol. 6, Sep. 2015, Art. no. 8264.
[12] T. Deffieux, Y. Younan, N. Wattiez, M. Tanter, P. Pouget, and
J. F. Aubry, “Low-intensity focused ultrasound modulates monkey
Visuomotor behavior,” Current Biol., vol. 23, no. 23, pp. 2430–2433,
2013.
[13] G. Li et al., “Improved anatomical specificity of non-invasive neuro-
stimulation by high frequency (5 MHz) ultrasound,” Sci. Rep., vol. 6,
p. 24738, Apr. 2016.
[14] S. S. Yoo et al., “Focused ultrasound modulates region-specific brain
activity,” Neuroimage, vol. 56, no. 3, pp. 1267–1275, 2011.
[15] Y. Tufail et al., “Transcranial pulsed ultrasound stimulates intact brain
circuits,” Neuron, vol. 66, no. 5, pp. 681–694, 2010.
[16] W. J. Tyler, Y. Tufail, M. Finsterwald, M. L. Tauchmann, E. J. Olson, and
C. Majestic, “Remote excitation of neuronal circuits using low-intensity,
low-frequency ultrasound,” PloS one, vol. 3, no. 10, p. e3511, 2008.
[17] M. D. Menz, Ö. Oralkan, P. T. Khuri-Yakub, and S. A. Baccus, “Precise
neural stimulation in the retina using focused ultrasound,” J. Neurosci.,
vol. 33, no. 10, pp. 4550–4560, 2013.
[18] O. Naor, Y. Hertzberg, E. Zemel, E. Kimmel, and S. Shoham, “Towards
multifocal ultrasonic neural stimulation II: Design considerations for an
acoustic retinal prosthesis,” J. Neural Eng., vol. 9, no. 2, p. 026006,
2012.
[19] “Food and Drug Administration (FDA). Guidance for industry and FDA
staff——Information for manufacturers seeking marketing clearance
of diagnostic ultrasound systems and transducers,” U.S. Dept. Health
Human Services, FDA, Center Devices and Radiologic Health, Tech
Rep., 2008.
View publication stats
1611
[20] R. L. King, J. R. Brown, W. T. Newsome, and K. B. Pauly, “Effective
parameters for ultrasound-induced in vivo neurostimulation,” Ultrasound
Med. Biol., vol. 39, no. 2, pp. 312–331, 2013.
[21] Y. Tufail, A. Yoshihiro, S. Pati, M. M. Li, and W. J. Tyler, “Ultrasonic
neuromodulation by brain stimulation with transcranial ultrasound,”
Nature Protocols, vol. 6, no. 9, pp. 1453–1470, 2011.
[22] A. Casper, D. Liu, and E. S. Ebbini, “Realtime control of multiple-
focus phased array heating patterns based on noninvasive ultrasound
thermography,” IEEE Trans. Biomed. Eng, vol. 59, no. 1, pp. 95–105,
Jan. 2012.
[23] E. S. Ebbini and C. A. Cain, “Multiple-focus ultrasound phased-array
pattern synthesis: Optimal driving-signal distributions for hyperthermia,”
IEEE Trans. Ultrason., Ferroelect., Freq. Control, vol. 36, no. 5,
pp. 540–548, Sep. 1989.
[24] A. Weyns, “Radiation field calculations of pulsed ultrasonic transducers:
Part 1: Planar circular, square and annular transducers,” IEEE Trans.
Ultrason., Ferroelect., Freq. Control, vol. 18, no. 4, pp. 183–188,
Jul. 1980.
[25] W. Legon et al., “Transcranial focused ultrasound modulates the activity
of primary somatosensory cortex in humans,” Nature Neurosci., vol. 17,
no. 2, pp. 322–329, 2014.
[26] H. Kim, A. Chiu, S. D. Lee, K. Fischer, and S. S. Yoo, “Focused
ultrasound-mediated non-invasive brain stimulation: Examination of
sonication parameters,” Brain Stimul., vol. 7, no. 5, pp. 748–756, 2014.
[27] A. C. Ho et. al., “Long-term results from an epiretinal prosthe-
sis to restore sight to the blind,” Ophthalmology, vol. 122, no. 8,
pp. 1547–1554, 2015.
[28] W. Lee, H. Kim, Y. Jung, I. U. Song, Y. A. Chung, and S. S. Yoo,
“Image-guided transcranial focused ultrasound stimulates human pri-
mary somatosensory cortex,” Sci. Rep., vol. 5, p. 08743, Mar. 2015.
[29] Q. Zhou, K. H. Lam, H. Zheng, W. Qiu, and K. K. Shung, “Piezoelectric
single crystal ultrasonic transducers for biomedical applications,” Prog.
Mater. Sci., vol. 66, pp. 87–111, Jun. 2014.
Authors’ photographs and biographies not available at the time of
publication..