Real-time Display and In-vivo Optical-resolution Photoacoustic

Microscopy for Ophthalmic Imaging

Sang-Won Lee

1, 2, 3

, Heesung Kang

and Tea Geol Lee

1, 2

Center for Nano-Bio Measurement, Korea Research Institute of Standards and Science, Daejeon, Rep. of Korea

Department of Nano Science, University of Science and Technology, Daejeon, Rep. of Korea

Center for Nanosafety Metrology, Korea Research Institute of Standards and Science, Daejeon, Rep. of Korea

Keywords: Optical-resolution Photoacoustic Microscopy, Ophthalmic Imaging, Angiography.

Abstract: Photoacoustic imaging is a non-invasive imaging technology that can be combined with optical absorption

contrast and detection of acoustic wave for structural, functional, and molecular imaging. Especially, optical-

resolution photoacoustic microscopy (OR-PAM) can provide a high spatial resolution with a micron-scale. In

this study, we have developed laser-scanning OR-PAM, which could obtain in-vivo photoacoustic ophthalmic

angiography. For high speed image acquisition, we used a nanosecond pulsed laser with a 300 kHz-pulse

repetition rates. In addition, we carried out parallel signal processing using a graphics processing unit to enable

fast signal processing. Therefore, we successfully obtained maximum amplitude projection images of

microvasculature in anterior and posterior segments of mouse’s eye with real-time display of 0.98 fps.

1 INTRODUCTION

Most of the imaging modalities in ophthalmology,

such as fundus camera (Pomerantzeff et al., 1979),

slit-lamp, scanning laser ophthalmoscope (Webb and

Hughes, 1981), and optical coherence tomography

(Huang et al., 1991), are based on detection of

reflected light or single-backscattered light (Jiao et

al., 2010). These imaging techniques have advantages

that can provide anatomical and functional images for

noninvasively accurate diagnosis of ocular diseases.

Optical coherence tomography (OCT) cannot provide

only three-dimensional and structural images but also

label-free functional images such as Doppler (Chen et

al., 1997), polarization (Ren et al., 2002), and

angiography (Wang et al., 2007). Additionally,

fundus camera and scanning laser ophthalmoscope

(SLO) can image retinal vasculature with high

transverse resolution using contrast agents such as

fluorescein and indocyanine green (Song et al., 2013).

However, these optical imaging tools cannot support

information of the molecules with optical absorption

properties in ocular tissues.

Photoacoustic imaging combined with optical

absorption contrast and detection of acoustic wave

has been actively studied because photoacoustic

imaging can provide structural, functional, and

molecular images (Wang et al., 2003, Zhang et al.

2006, Kim et al., 2010). Especially, optical-resolution

photoacoustic microscopy (OR-PAM) can provide

microscopic images with high spatial resolution using

tightly focused laser spot with a micro-scale (Maslov

et al. 2008, Xie et al. 2009)

Recently, several groups have developed

ophthalmic photoacoustic microscopy (Jiao et al.,

2010, Hu et al., 2010, de la Zerda et al., 2010,

Silverman et al., 2010). Hu et al. and de la Zerda et al.

have deomnstrated ocular OR-PAM images with an

acquisition speed of a few hours owing to usages of a

low pulse repetition rate laser and mechanical

scanning (Hu et al., 2010, de la Zerda et al. 2010).

Silverman et al. could achieve the imaging speed with

2.7 s for volumentric acqusition (Silverman et al.,

2010). However, their image pixel size was limited to

be 256  256 pixels. The imaging speed in OR-PAM

depends on the pulse repetition rate of the laser and

an acquired image pixel size.

The higher acquisition speed for an ophthalmic

image is important to reduce motion artifacts by

breathing, heartbeat, and eye movements. In addition,

small image pixel size can cause an low image pixel

resolution at the large field of view (FOV) although

an optical resolution by a tightly focused beam size is

very small.

Lee S., Kang H. and Geol Lee T.

Real-time Display and In-vivo Optical-resolution Photoacoustic Microscopy for Ophthalmic Imaging.

DOI: 10.5220/0006154300340038

In Proceedings of the 10th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2017), pages 34-38

ISBN: 978-989-758-215-8

In our previous study, we have demonstrated

maximum amplitude projection (MAP) images of

blood vessels in a mouse’s ear from volumetric data

set with the data size of 736  500  500 points at 1.02

seconds (Kang et al., 2015). For high speed image

acquisition, we used a nanosecond pulsed laser with

a pulse repetition rate of 300 kHz and carried out

parallel signal processing with graphics processing

unit (GPU).

In this study, we modified and applied our high-

speed laser-scanning OR-PAM system to obtain

photoacoustic ophthalmic images. Therefore, we

could obtain MAP images of microvasculature in

anterior and postierior segments of mouse’s eye with

real-time display of 0.98 fps and large pixel size of

500  500 pxiels.

2 EXPERIMENTAL METHODS

2.1 Laser-Scanning OR-PAM

Figure 1 shows a schematic diagram of a laser-

scanning OR-PAM. This schematic was modified

from our previous study (Kang et al., 2015). We used

an Ytterbium-doped fiber laser (YLP-G10, IPG

Photonics Corp.) with 1-ns pulse width and 300-kHz

pulse repetition rates at 532 nm. Light from the laser

was delivered by an optical fiber and scanned by 2-D

galvanometer scanning mirror with silver coating

(GVSM002, Thorlabs Inc.) as shown in Fig. 1 (a). In

our previous study, collimated light was scanned and

focused in inverse direction (bottom-to-top). In

addition, the acoustic waves was detected in direction

pass through a sample with optical path as shown in

Fig. 1(b). This schematic could obtain OR-PAM

images of blood vessels in mouse’s ear because

thickness of mouse’s ear is thin below 500 m.

However, in the eye, we could not apply acoustic

wave detection method in transmitted direction.

Therefore, we used a schematic of Fig. 1 (c) that the

acoustic waves occurred at the focal plane were

reflected by a thin glass with a tilting angle of 45 in

a water tank. Reflected acoustic wave detected by

ultrasound transducer. Finally, detected signals were

amplified by a pulser/receiver (5072PR, Olympus-

NDT) and digitally converted by a high-speed

digitizer (ATS9350, AlazarTech) at a sampling rate

of 250 MSamples/s.

For fast signal processing and real-time display,

we carried out parallel signal processing using GPU.

A graphics card (ASUS GTX780Ti, ASUSTeK

Computer Inc.) for GPU processing has 2880 stream

processors, a 7000 MHz memory clock and 3 GB of

RAM. To accelerate signal processing time and

display real-time OR-PAM images, we developed

custom software using Visual C++ of Visual Studio

2012 (Microsoft) and compute unified device

architecture (CUDA) technology (NVIDIA Corp.).

Figure 1: (a) Schematic diagram of laser-scanning OR-PAM for ophthalmic imaging, (b) previous our setup for light

illumination and acoustic wave detection, (c) modified setup for light illumination and acoustic wave detection.

Real-time Display and In-vivo Optical-resolution Photoacoustic Microscopy for Ophthalmic Imaging

2.2 Animal Preparation

We used BALB/c mice at 6 ~ 8 weeks of age as an

animal experiment model. The mice were housed

under standard conditions (room temperature 232°C,

humidity 50±10%) with a 12-h dark-light cycle and

were fed standard laboratory chow and water ad

libitum. The care, use, and interventions were

approved by the Korea Research Institute of

Bioscience and Biotechnology (KRIBB). Before

retinal imaging, mydriatic was applied to dilate the

pupil. During imaging, the anesthetized mice were

restrained in a customized holder.

3 RESULTS AND DISCUSSION

The volumetric data set was obtained at the laser

pulse repetition rate of 300 kHz and composed with

736 points per line, 500 lines per B-mode-frame, and

500 B-mode-frames per C-mode-frame. Therefore,

data size per volume was 351 MB (2 bytes  736 

500  500). In previous study, when we carried out

signal processing without parallel processing using

GPU, the total processing time was approximately

31.2 s. However, we could accelerate the processing

time to 1.02 s using parallel processing using GPU.

To evaluate the transverse resolution of our OR-

PAM system, we obtained the MAP image of a USAF

1951 resolution target (Edmunds Optics) as a sample.

When an area 3 mm  3 mm was achieved as the

maximum FOV, the lines at group 5 and element 6

could be distinguished. Therefore, a lateral resolution

was measured to be approximately 17.5 m. In

addition, when the FOV was reduced to 1 mm  1 mm,

we could distinguish the lines at group 7 and element

1, corresponding to a lateral resolution of 7.8 m

(Kang et al., 2015). This difference of the lateral

resolution came results from changing an image pixel

resolution owing to sizes of FOV rather than a spot

size of focused beam.

Figure 2 shows the MAP image (a) and 3-D

volumetric rendering image (b) of microvasculature

in the iris of a BALB/c mouse. When the

microvasculature image was obtained, a focused

ultrasound transducer at the center frequency of 20

MHz (V317-SU-F1.00in-PTF, Olympus-NDT) was

used. Focused ultrasound transducer had a 6 dB

bandwidth of 42.21%. In the anterior segment, the

vessels in the iris can be seen clearly in the MAP

image as sown in Fig. 2 (a). In addition, the location

of the pupil was well displayed as a hole at the center

Figure 2: In-vivo photoacoustic ophthalmic angiography of

the iris microvasculature of a BALB/c mouse. (a) MAP

image and (b) 3-D rendering image.

of the iris. Figure 2 (b) was processed with Amira 6.1

(FEI Corp.).

Figure 3 shows MAP images of retinal

vasculature (a) and sclera choroidal vasculature (b),

respectively. We used an unfocused ultrasound

transducer at the center frequency of 15 MHz when

the vasculature images in the posterior segment of the

mouse’s eye was acquired. In previous paper, light

with a pulse energy of 500 nJ was illuminated to

image a posterior eye using focused laser beam (Wu

et al., 2014). In this study, we used a pulse energy of

approximately 600 nJ. As shown in Fig. 3 (a), we

could clearly observed that blood vessels in the retina

were gathered into optic disk area (red arrow). When

the focal position of light was shifted onto deeper area

and the incident angle of light was adjusted, we could

obtain the choroidal vascular (red arrows) image in

BIOIMAGING 2017 - 4th International Conference on Bioimaging

Figure 3: In-vivo photoacoustic ophthalmic angiography of

the posterior segment of a BLAB/c mouse. (a) Retinal

vasculature and (b) sclera choroidal vasculature.

mouse’s sclera as shown in Fig. 3 (b).

PAM have been studied as useful molecular

imaging tool with contrast agents in various medical

fields. PAM will be also used as a preclinical imaging

tool in ophthalmology for drug development and

diagnosis of disease targeted with specific receptors

such as the vascular endothelial growth factor using

nanoparticles or dyes. In addition, if PAM is

combined with various ophthalmic imaging tools

(OCT, fundus, and SLO), we can obtain structural,

functional, and molecular information.

4 CONCLUSIONS

In conclusion, we demonstrated real-time display

photoacoustic ophthalmic angiography using laser-

scanning OR-PAM at a mouse’s anterior and

posterior segment. We could display MAP images

with 500  500 pixels as volumetric images at 0.98

fps when we used a nanosecond pulse laser with 300-

kHz pulse repetition rates. In further study, we will

obtain molecular images to apply diagnosis of ocular

disease using bio-conjugated contrast agents, which

are based on optical absorbance such as nanoparticles

and dyes.

ACKNOWLEDGEMENTS

This work was supported by the “Development of

Platform Technology for Innovative Medical

Measurement Program (KRISS-2016-16011064)”

from the Korea Research Institute of Standards and

Science. It was also supported by grants from the

“Pioneer Research Center Program (2012-0009541)”

and the “Nano Material Technology Development

Program (2014M3A7B6020163)” through the

National Research Foundation (NRF), Rep. of Korea.

REFERENCES

Chen, Z., Milner, S. S., Wang, X., Malekafzali, A., van

Germent, M. J. C., & Nelson, J. S., 1997. Noninvasive

imaging of in vivo blood flow velocity using optical

Doppler tomography. Optics Letters, 22, 1119-21.

de la Zerda, A., Paulus, Y. M., Teed, R., Bodapati, S.,

Dollberg, Y., Khuri-Yakub, B. T., Blumenkranz, M. S.,

Moshfeghi, D. M. & Gambhir, S. S., 2010.

Photoacoustic ocuar imaging. Optics Letters, 35, 270-

Hitenberger, C. K., Götzinger, E., Stricker, M., Pircher, M.,

& Fercher, A. F., 2001. Measurement and imaging of

birefringence and optic axis orientation by phase

resolved polarization sensitive optical coherence

tomography, Optics Express, 9, 780-90.

Hu, S. Rao, B., Maslov, K. & Wang, L. V., 2010. Label-

free photoacoustic ophthalmic angiography. Optics

Letters, 35, 1-3.

Huang, D., Swanson, E. A., Lin, C. P., Schuman, J. S.,

Stinson, W. G., Chang, W., Hee, M. R., Flotte, T.,

Gregory, K., Puliafito, C. A. & Fujimoto, J. G., 1991.

Optical coherence tomography. Science, 245, 1178-81.

Jiao, S., Jiang, M. Hu, J., Fawzi, A., Zhou, Q., Shung, K.

K., Pulifafito, C. A. & Zhang, H. F., 2010.

Photoacoustic ophthalmoscopy for in vivo retinal

imaging. Optics Express, 18, 3967-72.

Kang, H., Lee, S. W., Lee, E. S., Kim, S. H. & Lee, T. G.,

2015. Real-time GPU-accelerated processing and

volumetric display for wide-field laser-scanning

optical-resolution photoacoustic microscopy.

Biomedical Optics Express, 6, 4650-60.

Real-time Display and In-vivo Optical-resolution Photoacoustic Microscopy for Ophthalmic Imaging

Kim, C., Cho, E. C., Chen, J. Song, K. H., Au, L., Favazza,

C., Zhang, Q., Cobley, C. M., Gao, F., Xia, Y. & Wang,

L. V., 2010. In vivo molecular photoacoustic

tomography of melanomas targeted by bioconjugated

gold nanocages. ACS Nano, 4, 4559-64.

Maslov, K., Zhang, H. F., Hu, S. & Wang, L. V., 2008.

Optical-resolution photoacoustic microscopy for in

vivo imaging of single capillaries. Optics Letters, 33,

929-31.

Pomerantzeff, O., Webb, R. H. & Delori, F. C., 1979. Image

formation in fundus cameras. Invest Ophthalmol Vis Sci,

18, 630-7.

Silverman, R. H., Kong, F., Chen, Y. C., Lloyd, H. O., Kim,

H. H., Cannata, J. M., Shung K. K. & Coleman, D. J.,

2010. High-resolution photoacoustic imaging of ocular

tissues. Ultrasound in Medicine &Biology, 36, 733-42.

Song, W., Wei, Q., Feng, L. Sarthy, V. Jiao, S. Liu, X. &

Zhang, H. F., 2003. Multimodal photoacoustic

ophthalmoscopy in mouse, Journal of Biophotonics, 6,

505-12.

Wang, R. K., Jacques, S. L., Ma, Z., Hurst, S. Hanson, S. R.

& Gruber, A. Three dimensional optical angiography.

Optics Express, 15, 4083-97.

Wang, X., Pang, Y., Ku, G., Xie, X. Stoica, G. & Wang, L.

V., 2003. Noninvasive laser-induced photoacoustic

tomography for structural and functional in vivo

imaging of the brain. Nature Biotechnology, 21, 803-6.

Webb, R. H. & Hughes, G. W., 1981. Scanning laser

ophthalmoscope. IEEE Trans Biomed Eng, 28, 488-92.

Wu, N., Ye, S., Ren, Q. & Li, C., 2014. High-resolution

dual-modality photoacoustic ocular imaging. Optics

Letters, 39, 2451-4.

Xie, Z., Jiao, S., Zhang, H. F. & Puliafito, C. A., 2009.

Laser-scanning optical-resolution photoacoustic

microscopy. Optics Letters, 34, 1771-3.

Zhang, H. F., Maslov, K., Stoica, G. & Wang, L. V., 2006.

Functional photoacoustic microscopy for high-

resolution and noninvasive in vivo imaging. Nature

Biotechnology, 24, 848-51.

BIOIMAGING 2017 - 4th International Conference on Bioimaging