Open Access
Add to BibSonomy
Abstract
Understanding aqueous outflow resistance at the level of aqueous veins has been a challenge to the management of glaucoma. This study investigated resolving the anatomies of aqueous veins and the textures of surrounding sclera using photoacoustic microscopy (PAM). A dual wavelength PAM system was established and validated using imaging phantoms, porcine and human globes perfused with an optical contrast agent ex vivo. The system shows lateral resolution of 8.23 µm and 4.70 µm at 1200 nm and 532 nm, respectively, and an axial resolution of 27.6 µm. The system is able to separately distinguish the aqueous veins and the sclera with high contrast in full circumference of the porcine and human globes.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Glaucoma remains a leading cause of irreversible blindness worldwide [1,2]. Understanding outflow resistance distal to the trabecular meshwork is of great importance to minimally invasive glaucoma surgeries (MIGS) [3]. In the human eye, these distal tissues include Schlemm’s canal, the collector channels, and the aqueous veins that reside within the perilimbal sclera [4]. The roles of Schlemm’s canal and the collector channels on intraocular pressure (IOP) regulation has been extensively investigated [4]. Recent studies showed that episcleral veins and the aqueous veins possess filtration area comparable to that of the inner wall of Schlemm’s canal [5]. This finding is consistent with clinical experience that resistance at the level of aqueous veins may contribute to lack of IOP reduction after MIGS [3]. Aqueous veins reside in the perilimbal sclera [4,6]. The deformation of aqueous veins may be affected by the perilimbal sclera [7]. Therefore, resolving the aqueous veins and perilimbal sclera is a critical step towards understanding their roles in the outflow pathway and the mechanisms of flow resistance in the veins.
Conventional optical microscopy has limited tissue penetration and therefore cannot resolve the vasculature within the sclera. Aqueous angiography, using only two-dimensional (2D) imaging, has been employed to examine the functionality of distal vasculature to guide MIGS [8] and study the anatomies of the aqueous veins [5]. Optical coherence tomography (OCT) and OCT angiography (OCTA) have the ability to visualize the vasculature in the anterior segment of the eye, including aqueous veins [9–11]. In OCT images of perfused research eyes, the vasculature appears as void regions within the sclera [9]. Following empirical adjustments of the image contrast in the A-scans and complicated post-processing methods, the vessel contours of the aqueous veins may be extracted [10]. Significant segmentation artifacts have been observed in OCT imaging [12]. OCT angiography possesses unique advantage in resolving vasculature where active flow of optical scatterers, i.e., red blood cells, are present [12]. However, the aqueous veins that are proximal to the collector channels that emanate from Schlemm’s canal are relatively acellular as these veins are conduits for the aqueous humor from the anterior chamber and the trabecular meshwork (TM) or conventional outflow pathway. In addition, the slow physiological flow speed in aqueous veins also limits the accuracy in defining the anatomies. Finally, differentiating among episcleral, intrascleral, and aqueous veins requires empirical thresholding [9] and, therefore, involves uncertainty when aqueous veins are of particular interest.
Photoacoustic (PA) imaging is a spectrum of non-invasive and non-ionizing imaging technologies that rely on the optical absorption instead of scattering contrast or active flow [13]. The acoustic penetration in biological tissue allows PA imaging to recover anatomies under a translucent tissue surface [13]. Taking advantage of the unique optical absorption profiles of tissue components and contrast agents, multispectral PA imaging is able to capture multiple anatomies in parallel [14]. PA imaging has resolved multiple ocular anatomies [15], including the retinal pigment epithelium [16], vasculature at the anterior [7,17] and posterior [18] segments of the eye, especially in the lamina cribrosa [19],, intraocular tumors [20] and the whole globe [21,22], in both animal and human globes.
Using focused optical illumination, PA microscopy (PAM) is a promising tool to resolve the aqueous veins at high spatial resolution. A previous study by other researchers has resolved the vasculature at the anterior segments of mouse eyes using PAM [17]. Our study shows the feasibility of imaging the anatomy of aqueous veins, within a limited region, by perfusing the aqueous outflow pathway with contrast agent [7]. With the purpose of resolving the anatomies of aqueous veins and the textures of the surrounding perilimbal sclera around the full circumference of the cornea, this study, for the first time, developed and validated a multi-wavelength PAM system in both human and porcine eyes ex vivo. The imaging system will establish a platform for understanding the biomechanical interactions between the aqueous veins and the perilimbal sclera, and their roles in IOP regulation.
2. Method
2.1 Experiment setup
Figure 1. displays the diagram of the dual-wavelength photoacoustic microscopy system. A pulsed green fiber laser (GLPM-10, IPG Photonics, duration 1.2 ns, power 10 W, emission at 532 nm, maximum repetition rate 1 MHz) was employed to provide pulses at the wavelength of 532 nm. A fiber laser with emission at 1064 nm and pulse duration of 8 ns (YLPN-1-4 × 200-30-M, IPG Photonics, adjustable pulse duration, power 50 W, emission at 1064 nm, maximum repetition rate 1 MHz) was tuned to 1200nm-centered emission using a 60 m-long polarization maintaining single mode fiber (F-PM980, Newport) and a long pass filter (cutoff wavelength 1150 nm). The wavelength tuning is based on the principle of Raman scattering, as previously demonstrated by another research group [23,24]. The distribution of the wavelength after passing through the long pass filter was verified using a spectrometer. The two lasers were set to the same repetition rate at 125 KHz. These two wavelengths were chosen to target the characteristic optical absorption of the red contrast agent perfused in aqueous veins, and the collagen content in sclera with rich C-H chemical bonds [25], as shown later in the results section. The beams with two optical wavelengths were aligned coaxially as shown in Fig. 1. Then, the light beams went through a spatial filter to obtain two Gaussian shape beams with diameters of 7 mm. Following coaxial collimation and beam reshaping, light beams at the two wavelengths were focused on the sample surface by a scan lens (focal length 36 mm, LSM03-BB, Thorlabs, Inc., Newton, NJ) and scanned in two dimensions by a galvo mirror. During the scanning, light pulses at the two wavelengths were triggered in turns with 3 µs intervals. The laser energy at the focal plane was 80 nJ per pulse for 532 nm and 400 nJ per pulse for 1200 nm. The optical energy at 532 nm followed that used for retinal imaging in previous studies [26,27] under the optical energy safety limited established by American National Standard Institute (ANSI). The energy level at 1200 nm was set at 5 times that at 532 nm, which also agreed with the ANSI safety limit [28]. In addition, our previous study showed that the optical energy can be further reduced with decent imaging quality [29,30]. The sample holder has height adjustment, which ensured that the surface of the sample was always at the focal plane.
Fig. 1. Integrated dual-wavelength PAM system, LPDM: long pass dichroic mirror (cut-off wavelength 1150 nm), SPDM: short pass dichroic mirror (cut-off wavelength 900 nm), SL: scan lens. The green light pass is at the wavelength of 532 nm while the red-light path is centered at the wavelength of 1200 nm.
PA signals were captured using a needle hydrophone (Optosonic Inc., Arcadia, CA, center frequency: 27 MHz, 60% bandwidth) and amplified with 70 dB gain through a two-stage amplification circuit (Pasternack, 10-1000 MHz CA). The sensing surface was positioned approximately 1 mm away from the scanned surface and oriented toward the scanned area. The acoustic coupling was achieved by a few drops of transparent eye lubricant (GONAK, 2.5% sterile hypromellose ophthalmic demulcent solution). The lubricant generates ignorable PA signals at both wavelengths compared with the tissue signals. PA signals were converted to digital signals using a data acquisition card (CSE161G4, Gage, Vitrek, LLC 2022, IL) at a sampling rate of 250 MHz and stored in a computer. During the procedures, a digital delay/pulse generator (DG535, Stanford Research Systems, Sunnyvale, CA) synchronized the pulsed lasers, galvo mirrors, and data acquisition card. For displaying purpose, maximum amplitude mapping along the depth direction was used to covert 3D data into 2D images.
2.2 System resolution characterization
The lateral resolution of the system was measured by imaging the edge of a piece of black tape within an area of 0.5 mm x 0.5 mm at the step size of 1 µm. The edge spread function (ESF) and the corresponding point spread function (PSF), i.e. the spatial derivatives of the ESF were extracted from each scanning line across the tape edge in the image. The lateral resolutions were determined as the averaged FWHM quantified from all the lines.
The axial resolution was measured using a phantom containing randomly distributed microspheres. The microspheres have diameters of 6 µm (07312-5, Polybead Microspheres, Warrington, PA), with an approximate concentration of 30,000 per cubic millimeter. The background material of the phantom was 10% porcine gelatin (G2500, Sigma-Aldrich Co. LLC., Burlington, MA) with water. Using on our previous method, the 6 µm microspheres were uniformly and randomly distributed in 10% porcine gel [31,32]. We determined the axial resolution by slightly moving the phantom in axial dimension. The distance where the FWHM of the microsphere at two locations can be separated was determined as the axial resolution.
2.3 System validation with imaging phantoms
The reliability of differentiating veins perfused with red contrast agent and the surrounding tissue components using the dual-wavelength system was examined with an imaging phantom. Similar to the phantom described above, the imaging phantom was made of porcine gelatin dissolved in water at 10% by weight. Although gelatin also contains collagen, i.e. C-H bond, the concentration is much lower than that in the fishing line. As shown later in the results section, the background signal generated by the gelatin content is negligible. The phantom contained a hair filament, a fishing line, and a tunnel filled with red ink diluted to 5% of its original concentration (Kuro Sumi Corp., Fort Mill, SC). We measured the optical absorption spectrum of the red ink using a spectrophotometer (Multiskan SkyHigh Microplate Spectrophotometer A51119700DPC, Thermo Fisher Scientific Inc., Waltham, MA). The red ink strongly absorbs green light at 532 nm, as shown later in the results section. The fishing line is made of polyethylene. Similar to collagen, polyethylene has rich content of C-H chemical bonds [33] and strongly absorbed optical energy at 1200 nm. The melanin content in the hair filament absorbs both wavelengths.
2.4 Eye preparation
After receiving the eyes, all fat, extraocular muscles, and peribulbar tissues surrounding the sclera were carefully removed according to our previous procedures [7]. Before the experiment, the eye was kept hydrated by being surrounded with BSS-soaked tissues. Both human and porcine eyes were imaged 48 hours post-mortem. During the experiment, the eyes were placed in a cup-shaped holder with circulating water maintained at 37 °C. Such configuration mimics the core temperature of human body and the exposed anterior segment of the eye. Red ink was diluted to 5% by balanced salt solution (BSS). The red ink was gently injected into the posterior chamber of eyes, so it perfuses through the aqueous outflow pathway. Before the injection, and after the syringe needle was inserted into the eye, the needle tip was used to scratch around the inner angle of the globes to make an opening through the trabecular meshwork. This process reduced the resistance of ink perfusion into the aqueous veins.
2.5 Eye image acquisition
Dual wavelength images were acquired at 3 mm × 3 mm neighboring patches around the cornea and combined into a full circumferential, 3-D presentations of the aqueous veins and the surrounding perilimbal sclera. The eye holder was attached to a ball-joint and translational stage to maintain the horizontal orientation of the scanned surface. Each scanned patch included 500 × 500 lateral raster scanning steps and 2048 samples in the axial dimension. No Averaging was performed. The laser repetition rate was set at 125 KHz, allowing an image acquisition time of approximately 2 seconds per patch. The patches were combined by matching the overlapping regions to form full circumferential images in MATLAB manually. For human eyes, a full circumferential image consists of 15 image patches. The larger porcine eyes commonly include 20 image patches. We employed the Jerman Enhancement Filter MATLAB toolbox to reduce the background noise [34].
3. Results
3.1 Performance of the imaging system
The black tape measurements were performed at both wavelengths. Figure 2 shows the examples of the ESF and corresponding PSF. The lateral resolution was determined as 4.7 µm and 8.23 µm at 532 nm and 1200 nm, respectively. Equation (1) describes the theoretical focused laser spot size of a gaussian beam [35].
where M2 is beam quality parameter, λ is the wavelength, f is the focus depth, d is the beam diameter before the scanning lens. Theoretical calculations suggest that, with the assumption of an idea Gaussian beam (M2 = 1) and a measured beam size of 7 mm, the laser system should achieve resolutions of 3.5 µm at 532 nm and 7.8 µm at 1200 nm according to Eq. (1). The slight differences between the measured and theoretical resolutions can be attributed to the imperfect beam profile and optics in the system.Fig. 2. Lateral resolution of dual wavelength PAM system. (a, c) The PAM images of black tape at 1200 nm and 532 nm, respectively. (b, d) Examples of the edge spread functions (ESF) and the corresponding point spread functions (PSF) along the dashed lines in (a,c). The arrows mark the FWHM of the PSF.
The axial resolution is mostly determined by the hydrophone, and therefore, was measured at only 532 nm. Figure 3(a) shows the PAM image of the phantom with microspheres. The FWHM of the 6 µm microsphere is 55.2 µm on average as displayed in Fig. 3(b, d). We determined the axial resolution by slightly moving the phantom in the axial dimension. The distance where the FWHM of the microsphere at two locations can be separated was determined as the axial resolution, as illustrated in Fig. 3(c), which is 27.6 µm.
Fig. 3. Axial resolution of dual wavelength PAM System. (a) The PAM image of 6 um diameter microsphere phantom. (b) The example PA signal of one microsphere. The red solid curve is the envelope of the original PA signal plotted in blue dashed curve. The black curve is the Gaussian fitted line of the envelope. (c) Determining the axial resolution. The dashed curves are PA signals generated by the same microsphere with a 27 µm shift along the axial dimension. The solid red line is the summed envelope of the two dashed signals. The signals at the two locations can be separated by their FWHM. (d) The FWHM distribution of microspheres in Fig. 3(a).
3.2 Phantom images
As expected, PAM at 532 nm only captures the red dye and hair fiber, whereas that at 1200 nm only captures the fish lines and the hair fiber in Fig. 4. This supports our hypothesis that the optical contrast agent facilitates distinguishing between the aqueous veins and the surrounding sclera anatomies.
Fig. 4. PAM images of the imaging phantom made of hair, fishing line and red ink. (a) Optical absorption profiles of red ink, melanin in hair fiber [36] and C-H chemical bonds in fishing line [25,33]. (b) The photograph of the three inclusions. (c) Images acquired at the wavelengths of 1200 nm and 532 nm, respectively.
3.3 Resolving the aqueous veins and perilimbal sclera in porcine and human eyes
Figure 5 shows dual-wavelength PAM images taken from the perilimbal region in an intact porcine eye. As shown in Fig. 5(a), our PAM system is able to acquire full circumferential anatomies of aqueous veins, as well as the texture of the surrounding sclera. Magnified images in Fig. 5(b) and (d) show that PAM at 532 nm can resolve the vasculatures with a large variation of diameters, ranging from approximately 100 to 300 micrometers. Despite the dimensional variations, the vasculature is resolved with relatively uniform contrast and low background noise. The images taken at 1200 nm within the same regions show the textures of the sclera with spatial patterns distinctive from those in the vasculature images. Representative images from another 2 porcine eyes are included in the supplemental material (Figure S.1 and S.2).
Fig. 5. Dual-wavelength PAM images of a porcine eye. In all panels, aqueous vasculature was rendered in red, and the textures of the surrounding sclera are rendered in gray. (a) Full circumferential image with both components. (b-c) and (d-e) are magnified regions A and B in (a), respectively, with the two tissue components displayed separately.
Figure 6 shows the images of an intact human globe. Similar to the porcine eye image, the anatomies of the aqueous veins and the scleral texture are separately resolved with uniform and distinctive contrasts over the background. Compared to the porcine eye, the diameters of the aqueous veins are finer, with approximate sizes of 50-100 micrometers. We further examined the system performance in resolving the spatial features in depth i.e., axial dimension. Figure 7 shows that the system can resolve the cross-section of the aqueous veins and the texture features in the sclera, as well as their locations in depth.
Fig. 6. Dual-wavelength PAM images of a human eye. (a) Full circumferential image with both components. The magnified features of vasculature and sclera tissues of the area A and B in the white boxes in (a) are shown in (b-c) and (d-e), respectively.
Fig. 7. Resolving the depths of spatial features in a human eye. (a) Aqueous veins image acquired at 532 nm. The depths of individual vessels were encoded in color scale. (b-c) Cross-sectional images of the sclera texture and aqueous veins along the yellow dotted line in (a), respectively.
4. Discussion
The PAM system established in this study successfully resolved aqueous veins and the surrounding scleral texture. Relying on characteristic optical absorption instead of scattering, PAM possesses unique advantages in revealing vasculatures with high and uniform contrasts over background signals and noises. The contours of the vasculature can therefore be precisely extracted without post-processing approaches that introduces uncertainties as found in OCT images [9]. As mentioned in the introduction section, aqueous veins have slow flow rate by its intrinsic physiology [37,38]. Optical absorption-based PAM does not require active flow and the contrast is not correlated with the flow speed. Therefore, PAM may be an ideal tool for the investigation of the outflow mechanisms at the level of aqueous veins.
Contrast agent is required for the study of aqueous vein using PAM, as the veins are acellular or contain very few red blood cells [39]. In this study, we used red ink with particle size ranging from hundreds of nm to hundreds of µm. In addition to the anatomies, such contrast agent also allows for the measurement of flow rate within each vessel, as investigated in our previous study [40]. Molecular imaging contrast agents, such as ICG and fluorescein, may also be used. These contrast agents have been commonly used in clinical procedures [8] and have greater potential for clinical translation.
Currently our PAM system utilizes a needle hydrophone that receives the PA signals at a fixed angle slightly off from the axial direction of the illumination beam. The receiving angle was compensated for in all the 3D images. In the future, we will upgrade our system with a coaxial imaging geometry [41,42], where the receiving aperture of an ultrasound traducer is along with the scanning beams using a 45-degree acoustic reflector. Such configuration, according to previous works [41,42], will improve the imaging penetration and axial resolution. The lateral scanning step we employed was approximately 6 µm, which was slightly larger than the lateral resolution of 4.7 µm at 532 nm. As shown in the results section, the aqueous veins have diameters of 100-300 microns in porcine eyes and 50-100 microns in human eyes. We selected this scanning step size to balance the number of scanning patches around the eye and ability to clearly resolve the vasculature.
Our previous studies with single illumination wavelength have shown local deformation of the aqueous veins as a function of IOP regulation. With the ability to acquire full circumferential anatomical information mentioned above, and by integrating the flow measurements in the future, the PAM system in this study will establish a flatform for investigating the resistance of aqueous outflow pathway at the level of aqueous veins.
5. Conclusion
This study established a dual-wavelength PAM for resolving the anatomies of aqueous veins and textures of the surrounding sclera. The system achieved lateral resolutions of 8.23 µm and 4.70 µm at 532 nm and 1200 nm, respectively, and an axial resolution of 27.5 mm. The system was successfully validated with porcine and human globes ex vivo.
Funding
Glaucoma Foundation; National Institute of Diabetes and Digestive and Kidney Diseases (1R01DK12568701); National Cancer Center (5R37CA22282903); National Eye Institute (5P30EY007003); National Science Foundation (CMMI1760291).
Disclosures
The authors declare no conflicts of interest.
Data Availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
Supplemental document
See Supplement 1 for supporting content.
References
1. S. R. Flaxman, R. R. A. Bourne, S. Resnikoff S. Vision Loss Expert Group of the Global Burden of Disease, et al., “Global causes of blindness and distance vision impairment 1990-2020: a systematic review and meta-analysis,” Lancet Global Health 5(12), e1221–e1234 (2017). [CrossRef]
2. H. A. Quigley and A. T. Broman, “The number of people with glaucoma worldwide in 2010 and 2020,” Br. J. Ophthalmol. 90(3), 262–267 (2006). [CrossRef]
3. P. Agrawal and S. E. Bradshaw, “Systematic Literature Review of Clinical and Economic Outcomes of Micro-Invasive Glaucoma Surgery (MIGS) in Primary Open-Angle Glaucoma,” Ophthalmol Ther 7(1), 49–73 (2018). [CrossRef]
4. T. Carreon, E. van der Merwe, R. L. Fellman, et al., “Aqueous outflow - A continuum from trabecular meshwork to episcleral veins,” Prog. Retinal Eye Res. 57, 108–133 (2017). [CrossRef]
5. E. D. K. Cha, J. Xu, L. Gong, et al., “Variations in active outflow along the trabecular outflow pathway,” Exp. Eye Res. 146, 354–360 (2016). [CrossRef]
6. M. Johnstone, C. Xin, J. Tan, et al., “Aqueous outflow regulation – 21st century concepts,” Prog. Retinal Eye Res. 83, 100917 (2021). [CrossRef]
7. L. Ni, J. Riesterer, H. Wang, et al., “Method for the biomechanical analysis of aqueous veins and perilimbal sclera by three-dimensional photoacoustic imaging and strain field calculation,” Sci. Rep. 11(1), 22108 (2021). [CrossRef]
8. T. Dada, S. Verma, A. N. Bukke, et al., “Aqueous Angiography-guided Minimally Invasive Glaucoma Surgery,” Journal of current glaucoma practice 16(1), 1–3 (2022). [CrossRef]
9. L. Kagemann, G. Wollstein, H. Ishikawa, et al., “3D visualization of aqueous humor outflow structures in-situ in humans,” Exp. Eye Res. 93(3), 308–315 (2011). [CrossRef]
10. A. W. Francis, L. Kagemann, G. Wollstein, et al., “Morphometric Analysis of Aqueous Humor Outflow Structures with Spectral-Domain Optical Coherence Tomography,” Invest. Ophthalmol. Visual Sci. 53(9), 5198–5207 (2012). [CrossRef]
11. S. E. Moroi, D. M. Reed, D. S. Sanders, et al., “Precision medicine to prevent glaucoma-related blindness,” Current opinion in ophthalmology 30(3), 187–198 (2019). [CrossRef]
12. A. Zhang, Q. Zhang, C.-L. Chen, et al., “Methods and algorithms for optical coherence tomography-based angiography: a review and comparison,” J. Biomed. Opt. 20(10), 100901 (2015). [CrossRef]
13. L. V. Wang and S. Hu, “Photoacoustic tomography: in vivo imaging from organelles to organs,” Science 335(6075), 1458–1462 (2012). [CrossRef]
14. B. Cox, J. G. Laufer, S. R. Arridge, et al., “Quantitative spectroscopic photoacoustic imaging: a review,” J. Biomed. Opt. 17(6), 061202 (2012). [CrossRef]
15. Z. Hu, Q. Liu, and Y. M. Paulus, “New frontiers in retinal imaging,” International Journal of Ophthalmic Research 2(3), 148–158 (2016). [CrossRef]
16. M. Jiang, X. Zhang, C. A. Puliafito, et al., “Adaptive optics photoacoustic microscopy,” Opt. Express 18(21), 21770–21776 (2010). [CrossRef]
17. S. Jeon, H. B. Song, J. Kim, et al., “In Vivo Photoacoustic Imaging of Anterior Ocular Vasculature: A Random Sample Consensus Approach,” Sci. Rep. 7(1), 4318 (2017). [CrossRef]
18. W. Au - Song, Q. Au - Wei, S. Au - Jiao, et al., “Integrated Photoacoustic Ophthalmoscopy and Spectral-domain Optical Coherence Tomography,” JoVE 1(71), e4390 (2013). [CrossRef]
19. T. Chuangsuwanich, M. Moothanchery, A. Tsz Chung Yan, et al., “Photoacoustic imaging of lamina cribrosa microcapillaries in porcine eyes,” Appl. Opt. 57(17), 4865–4871 (2018). [CrossRef]
20. G. Xu, Y. Xue, Z. G. Özkurt, et al., “Photoacoustic imaging features of intraocular tumors: Retinoblastoma and uveal melanoma,” PLoS One 12(2), e0170752 (2017). [CrossRef]
21. R. H. Silverman, F. Kong, Y. C. Chen, et al., “High-Resolution Photoacoustic Imaging of Ocular Tissues,” Ultrasound in Medicine & Biology 36(5), 733–742 (2010). [CrossRef]
22. A. de la Zerda, Y. M. Paulus, R. Teed, et al., “Photoacoustic ocular imaging,” Opt. Lett. 35(3), 270–272 (2010). [CrossRef]
23. T. Buma, B. C. Wilkinson, and T. C. Sheehan, “Near-infrared spectroscopic photoacoustic microscopy using a multi-color fiber laser source,” Biomed. Opt. Express 6(8), 2819–2829 (2015). [CrossRef]
24. D. Koeplinger, M. Liu, and T. Buma, “Photoacoustic microscopy with a pulsed multi-color source based on stimulated Raman scattering,” in 2011 IEEE International Ultrasonics Symposium, (2011), 296–299.
25. H.-W. Wang, N. Chai, P. Wang, et al., “Label-Free Bond-Selective Imaging by Listening to Vibrationally Excited Molecules,” Phys. Rev. Lett. 106(23), 238106 (2011). [CrossRef]
26. W. Liu, K. M. Schultz, K. Zhang, et al., “In vivo corneal neovascularization imaging by optical-resolution photoacoustic microscopy,” Photoacoustics 2(2), 81–86 (2014). [CrossRef]
27. C. Tian, W. Zhang, A. Mordovanakis, et al., “Noninvasive chorioretinal imaging in living rabbits using integrated photoacoustic microscopy and optical coherence tomography,” Opt. Express 25(14), 15947–15955 (2017). [CrossRef]
28. A. Standard, Z136. 1. American national standard for the safe use of lasers (American National Standards Institute, Inc., (1993).
29. Y. Li, W. Zhang, V. P. Nguyen, et al., “Retinal safety evaluation of photoacoustic microscopy,” Exp. Eye Res. 202, 108368 (2021). [CrossRef]
30. W. Zhang, Y. Li, V. P. Nguyen, et al., “Ultralow energy photoacoustic microscopy for ocular imaging in vivo,” J. Biomed. Opt. 25(06), 1 (2020). [CrossRef]
31. L. Ni, J. Siddiqui, A. M. Udager, et al., “Characterizing the aggressiveness of prostate cancer using an all-optical needle photoacoustic sensing probe: feasibility study,” Biomed. Opt. Express 12(8), 4873–4888 (2021). [CrossRef]
32. G. Xu, I. A. Dar, C. Tao, et al., “Photoacoustic spectrum analysis for microstructure characterization in biological tissue: A feasibility study,” Appl. Phys. Lett. 101(22), 1 (2012). [CrossRef]
33. R. Li, Justin Rajesh Rajian, Pu Wang, et al., “Vibrational Photoacoustic Tomography: Chemical Imaging beyond the Ballistic Regime,” J. Phys. Chem. Lett. 4(19), 3211–3215 (2013). [CrossRef]
34. T. Jerman, F. Pernus, B. Likar, et al., “Enhancement of Vascular Structures in 3D and 2D Angiographic Images,” IEEE Trans. Med. Imaging 35(9), 2107–2118 (2016). [CrossRef]
35. H. Urey, “Spot size, depth-of-focus, and diffraction ring intensity formulas for truncated Gaussian beams,” Appl. Opt. 43(3), 620–625 (2004). [CrossRef]
36. S. L. Jacques, “Optical properties of biological tissues: a review,” Phys. Med. Biol. 58(11), R37–R61 (2013). [CrossRef]
37. K. W. Ascher, “Aqueous veins and their significance for pathogenesis of glaucoma,” Arch. Ophthalmol. 42(1), 66–76 (1949). [CrossRef]
38. M. Johnson and R. Kamm, “The role of Schlemm's canal in aqueous outflow from the human eye,” Invest. Ophthalmol. Visual Sci. 24, 320–325 (1983).
39. K. W. Ascher, “The Aqueous Veins: I. Physiologic importance of the visible elimination of intraocular fluid,” Am. J. Ophthalmol. 192, xxix–liv (2018). [CrossRef]
40. S.-L. Chen, Z. Xie, P. L. Carson, et al., “In vivo flow speed measurement of capillaries by photoacoustic correlation spectroscopy,” Opt. Lett. 36(20), 4017–4019 (2011). [CrossRef]
41. J. Rebling, M. Ben-Yehuda Greenwald, M. Wietecha, et al., “Long-term imaging of wound angiogenesis with large scale optoacoustic microscopy,” Adv. Sci. 8(13), 2004226 (2021). [CrossRef]
42. W. Xing, L. Wang, K. Maslov, et al., “Integrated optical- and acoustic-resolution photoacoustic microscopy based on an optical fiber bundle,” Opt. Lett. 38(1), 52–54 (2013). [CrossRef]
