In-vivo functional and structural retinal imaging using multiwavelength photoacoustic remote sensing microscopy

Abstract

Many important eye diseases as well as systemic disorders manifest themselves in the retina. Retinal imaging technologies are rapidly growing and can provide ever-increasing amounts of information about the structure, function, and molecular composition of retinal tissue in-vivo. Photoacoustic remote sensing (PARS) is a novel imaging modality based on all-optical detection of photoacoustic signals, which makes it suitable for a wide range of medical applications. In this study, PARS is applied for in-vivo imaging of the retina and estimating oxygen saturation in the retinal vasculature. To our knowledge, this is the first time that a non-contact photoacoustic imaging technique is applied for in-vivo imaging of the retina. Here, optical coherence tomography is also used as a well-established retinal imaging technique to navigate the PARS imaging beams and demonstrate the capabilities of the optical imaging setup. The system is applied for in-vivo imaging of both microanatomy and the microvasculature of the retina. The developed system has the potential to advance the understanding of the ocular environment and to help in monitoring of ophthalmic diseases.

Figure 1

Retina imaging setup (a) Simplified schematic of the developed system. C: Collimator, Ci: Circulator, DC: Dispersion compensation, DM: Dichroic mirror, GVS: Galvo Scanner, L: Lens, P: Polarization controller, PBS: Polarized beam splitter, PD: Photodiode. QWP: Quarter wave plate, SF: Spectral Filter, SP: Short pass filter, VCSEL: Vertical Cavity Surface Emitting Laser. (b) Imaging optics arrangement of the telecentric pair for retina imaging. The distances between the elements are as F1 = 50 mm, F2 = 30 mm.

Figure 2

Imaging phantom eye models for human and rat. Simplified eye model consisting of the cornea, aqueous humor, pupil, crystalline lens, vitreous chamber, and the retina. (a) Human eye model (b). Strings of 7 μm carbon fibers are placed at the bottom of the eye model (c). CF image acquired with PARS scattering mechanism (d). CF image acquired using PARS absorption mechanism (e). Custom rat eye model consisting of a single achromatic lens and a 3D printed plastic chamber and strings of carbon fibers (f). PARS scattering and absorption contrast images, respectively (g, h). PARS excitation and detection beams are co-aligned so that the same CF are in focus in both images (white arrows).

Figure 3

In-vitro bovine blood phantom experiment. Experimental setup of the experiment (A). Images of glass capillary with flowing blood acquired at 532 nm and 558 nm excitation wavelengths, respectively (B, C).

Figure 4

Volumetric and cross-sectional OCT images. Cross-sectional images acquired in-vivo from rat retina showing distinct layers of the retina. CH: choroid, CRA: central retinal artery, INL: inner nuclear layer, IPL: inner plexiform layer, IS/OS junction of inner segment and outer segment layer, NFL: nerve fiber layer, ONL: Outer nuclear layer, ONH: optic nerve head, OPL: outer plexiform layer, RPE retinal pigment epithelium layer. (a, b). OCT fundus images visualizing optic nerve head, large retinal vessels, optic nerve fiber bundle (yellow arrows), deeper retinal layer microvasculature (red arrows) (c–e).

Figure 5

PARS retinal imaging. Fundus PARS image acquired from large vessels around ONH from a 2.6 × 2.6 mm area (a). zoomed-in section of one of the vessels acquired from a similar area with smaller vasculature (b). Fundus image acquired using scattering contrast of PARS system showing arteries (red arrows) and veins (blue arrows) (c). Oxygen saturation map in the retina obtained using multiwavelength PARS imaging (d).