This PDF file contains the front matter associated with SPIE Proceedings Volume 10858, including the Title Page, Copyright information, Table of Contents, Author and Conference Committee lists.

Diabetic retinopathy (DR) and other eye diseases can affect artery and vein differently. Therefore, differential artery-vein analysis can improve disease detection and treatment assessment. This study aims to establish color fundus image analysis guided artery-vein differentiation in OCTA, and to verify that differential artery-vein analysis can improve the sensitivity of OCTA detection and classification of DR. Briefly, optical density ratio (ODR) analysis and blood vessel tracking were combined to identify artery-vein in color fundus images. The fundus artery-vein map was used to register arteries and veins in corresponding OCTA images. Based on the fundus image guided artery-vein differentiation, quantitative analysis of arteries and veins in control and NPDR OCTA images were performed. The sensitivities of traditional mean blood vessel caliber (m-BVC) and artery-vein ratio of BVC (AVR-BVC) were quantitatively compared for DR classification. One way, multi-label analysis of variance (ANOVA) with Bonferroni’s test and Student t-test was employed for evaluating classification performance. Images from 20 eyes of 18 control subjects and 48 eyes of 35 NPDR patients (18 mild, 16 moderate and 14 severe NPDR) were used for this study. Compared to m-BVC, AVR-BVC provided enhanced sensitivity in differentiating NPDR stages. AVR-BVC was able to differentiate among control and three different NPDR groups. AVR-BVC could also differentiate control from mild NPDR, promising a unique OCTA biomarker for detecting early onset of NPDR.

Spontaneous retinal venous pulsations (SRVP) describe rhythmic caliber oscillations of one or multiple major retinal veins at the site of the optic nerve head (ONH). This phenomenon is reported to possibly enable non-invasive intracranial pressure (ICP) assessment besides its potential significance for major ocular diseases such as glaucoma or diabetic retinopathy. In this work, we illustrate the advantages of optical coherence tomography (OCT) imaging for investigation of SRVP. Using conventional intensity based OCT as well as the functional extension Doppler OCT (DOCT), the pulsatile changes in venous vessel caliber are analyzed qualitatively and quantitatively. Single-channel and double-channel line scanning protocols of our time-encoded multi-channel OCT prototype are employed to investigate venous caliber oscillations as well as venous flow pulsatility in the eyes of healthy volunteers. A comparison to recordings of scanning laser ophthalmoscopy – a standard en-face imaging modality for evaluation of SRVP – is provided, emphasizing the advantages of tomographic image acquisition. To the best of our knowledge, this is the first quantitative time-resolved investigation of SRVP and associated retinal perfusion characteristics using OCT.

Despite the recent development of advanced ophthalmic imaging techniques, volumetric fluorescence angiography (vFA) over a large field of view is still lacking. Fundus photography techniques have significant limitations due to the lack of 3D imaging capability. Scanning laser ophthalmoscopy (SLO) and confocal SLO (cSLO) use confocal gating to remove diffused light, resulting in crisper image quality. However, the volumetric imaging of SLO requires to compile z stacks, which can be challenging and time-consuming. Adaptive optics SLO (AOSLO) allows diffraction-limited resolution in both axial and lateral resolution. This technique is limited however, by its small field of view (FOV) and also the necessity of z stacks for volumetric imaging. To fill the technical void of vFA over a large field of view (FOV), we developed a novel retinal imaging modality called oblique scanning laser ophthalmoscopy (oSLO) for in vivo volumetric fluorescence retinal imaging. By using oblique illumination and detection, oSLO essentially allows “OCT-like” cross-sectional images contributed solely by the fluorescent contrast, without the need for z stacking. We will demonstrate 3D vFA over a 30˚x30˚ FOV in vivo in mouse retina. We will further report a high-speed oSLO in imaging capillary hemodynamics. The new capability allows the calculation of capillary hematocrit and blood speed in 3D, which can be potentially valuable in diabetic retinopathy and macular degeneration.

Vision is the most important sense organ of human, more than 80% of the information from outside world is acquired by vision. Vision starts at the photoreceptors in the retina capturing the visible light photons. There are two general types of photoreceptors, called rods and cones. Rods allow us to see in dim and dark light, cones allow us to perceive fine visual detail and color. To understand physiology of cones, researchers developed many model organisms that allow them to study in details different aspects of photoreceptors function. Specifically, mice play a central role in basic vision science research. However, one should keep in mind that mice have rod dominant retinas which is different from human cone dominant retinas near fovea. As one of the consequence in vivo imaging of cones in humans is relatively easy in periphery, and cone mosaic was the first cellular structure that was reported to be seen by optical coherence tomography (OCT) and scanning laser ophthalmoscopy (SLO), especially with implementation of adaptive optics (AO)[1]. However, just recently researchers started to visualize human rods which are smaller than cones [2, 3]. In case of mouse retinal imaging, it is quite the opposite situation. There have been recent reports of imaging rods mosaic [4-6], but up to date no reports on identifying cones in the images. Given that the cones are twice as big as rods in mice, it is very interesting why one can visualize rods but cannot visualize cones.

Leaky vasculature is a key feature in a number of retinal diseases such as diabetic retinopathy and age related macular degeneration and is commonly associated with neovascularization. Currently, the only way to identify leaky vasculature is through fluorescence angiography, which lacks depth resolution and the ability to precisely localize leaky vessels. Here we present the first 4D tracking of leaky vasculature in a mouse model of sub-retinal neovascularization using contrast-enhanced OCT. A very-low-density-lipoprotein receptor kockout mouse model was imaged with OCT angiography at multiple time points following intravenous injection of Intralipid 20%, an OCT contrast agent. Compared to healthy vessels, leaky vessels appeared to become broader over time. By fitting a model to mean intensity projection profiles, the apparent width of the vessels was quantified as an indicator of leakage. A clear trend of increased leakage following the injection of contrast was observed in vessels that derive from retinal lesions. This finding was likely caused by the infiltration of Intralipid particles into the surrounding retinal tissue. Intralipid is an ideal OCT contrast agent as it is FDA approved for human use as an intravenous nutritional supplement and is highly scattering, which makes it a strong candidate for future clinical translation. To summarize, we have demonstrated 4D tracking of individual leaky vessels for the first time using contrast-enhanced OCT in a mouse model of neovascularization. This technique improves upon the capabilities of fluorescence angiography and may help pave the way for clinical translation of contrast-enhanced OCT methods for enhanced diagnostic specificity.

Never before have optical, information, and biomedical technologies converged to the extent they do today. Moreover, the accessibility of enormous amounts of computing power at reasonable cost has transformed raw data into useable information that can enable biological discoveries and medical decisions to be made faster and more accurate than ever before. In the field of ophthalmology in particular, advanced optical instrumentation, devices, and procedures have revolutionized the standard of care and improved outcomes for millions, for the benefit of society. The cornea is the clear outer window of the eye and is directly accessible for examination and treatment using light-based approaches, and as such it provides us with a unique window into the physiology of the body in health and disease. At the same time, the cornea is a model tissue from which we have acquired much knowledge about light-tissue interactions. Finally, and importantly, diseases of the cornea compromising its transparency are responsible for millions of cases of corneal blindness globally, so there is much to gain from technological advancements in the field. Here, a need for new technological solutions is presented, that is not primarily technology-driven but instead motivated by real and pressing medical needs in the research, diagnosis and treatment of corneal blindness.

High speed optical coherence tomography (OCT) systems with A-scan rates greater than 100 kHz allow for 4D visualizations in applications such as intraoperative OCT. However, traditional triangle or sawtooth waveforms used to drive galvanometer scanners often have frequency content that exceeds the bandwidth of the scanners, leading to distorted scans. Sinusoidal waveforms used to drive resonant scanners also lead to distorted scans due to the nonlinear scan velocity. Additionally, with raster scan patterns, the scanner needs time to stop and reverse direction in between B-scans, leading to significant acquisition dead time. Continuous scan patterns such as constant frequency spiral scanning or Lissajous scanning no longer have acquisition dead times, but suffer from non-uniform sampling across the imaging plane. We previously introduced constant linear velocity (CLV) spiral scanning as a novel scan pattern to maximize the data acquisition efficiency of high speed OCT systems. While this continuous scan pattern has no acquisition dead time and produces more uniform sampling compared to raster scanning, it required significant processing time. We introduce a processing pipeline implemented using CUDA in C++, which drastically reduces the amount of processing time needed, allowing real time visualization of 4D OCT data. To demonstrate its potential utility, we used CLV scanning with a 100 kHz swept-source OCT system to image retinas of enucleated porcine eyes undergoing mock ophthalmic surgery movements. Additionally, we rendered these volumes in virtual reality (VR) in real time, allowing for interactive manipulation and sectioning.

Optical coherence tomography (OCT) is currently recognized as the gold standard for identifying retinal structural abnormalities in ophthalmology. However, its availability is often limited to large eye centers and research labs due to its high cost and lack of portability. We present a low-cost, portable spectral-domain OCT system with a total cost of materials under $6,000. Compared to current commercial systems, our design offers 50% size reduction and over 80% cost reduction. Image acquisition interface is incorporated and displayed onto a mounted 7-inch touchscreen. Human retinal imaging is demonstrated, and performance is compared with a commercial OCT system. Based on contrast-to-noise ratio analysis, the low-cost OCT demonstrates comparable imaging capabilities.

Imaging the entire human cornea with a conventional OCT system configuration requires trade-offs between resolution and depth-of-focus because the cornea is curved over a depth of approximately 4 mm. These system trade-offs result in image quality variations in the corneal image such as a bright apex surrounded by decreasing intensity as the cornea curves away from the apex. These intensity changes cause non-biological ambiguities in interpreting the image, make it difficult to see anatomy in the dim areas, and make automated surface detection difficult in the periphery. To address this problem, we developed a continuously focusing corneal OCT system coupled with a constant linear velocity (CLV) spiral scan pattern that is able to better maintain focus from the apex to the deeper cornea during a scan. The continuous focusing was implemented by introducing a focusing telescope on a motorized stage into the sample arm and matching the translation of the telescope with the CLV scan as it spiraled from the corneal apex outwards. Orthogonal B-scans prior to volume acquisition were used as a reference to estimate and correct motion that occurred during the subsequent CLV scan. A consented subject was imaged, and the resultant image showed increased intensity in the peripheral and deeper cornea and anterior chamber. Continuous focusing with CLV spiral scanning is a promising design change to OCT systems allowing adequate focus over relatively large depths such as for scanning the human cornea.

Anatomical changes of the growing crystalline lens influence its refractive development, including power and spherical aberration. We have recently developed a new instrument that characterizes both the optical and biometric properties of the lens in-vitro by merging Ray-Tracing Aberrometry (RTA) with three-dimensional OCT imaging. In this abstract, we describe the application of the RTA to the measurement of lens spherical aberration. Experiments were performed on 54 isolated human lenses (age: 0.25 to 56 years). The system was programmed to sequentially deliver the probing beam through the lens using a raster scan pattern of 13 × 13 transversal positions spaced 0.5 mm apart. Exit rays were imaged after exiting the tissue chamber at 9 different axial positions (ΔZ = 0 mm to 8 mm) in 1 mm intervals. A total of 1,521 spot images were acquired per lens. All data was automatically analyzed using custom software we developed in MATLAB. Exit ray slopes over a 6 mm pupil were used to determine Zernike wavefront coefficients up to the sixth order. The 4th order Zernike coefficient Z[4,0] was used to measure primary spherical aberration (SA). The results suggest that spherical aberration of the growing lens becomes more negative before adulthood and less negative after around age 30. The data is consistent with results from in-vivo studies that suggest the lens spherical aberration becomes less negative in older lenses (>30 years).

Our recent work has demonstrated the application of femtosecond oscillators to crosslink corneal collagen in absence of photosensitizers to correct refractive errors and enhance corneal mechanical properties. Here we propose a new design of the femtosecond laser station enabling significantly reduce the time of the treatment procedure to clinically acceptable duration and report the results of one-dimensional parametric study on the control of the adjustment of corneal curvature. Fresh porcine eyes were subject to the laser irradiation ex vivo. The volumetric exposure to the laser has been executed by treating multiple planar areas at varying depths, measured from the ocular surface. The degree of the refractive error correction was adjusted by varying the number the laser treated planar areas. The eye topography has been monitored within 24-hour window post-treatment to assess degree and short term stability of the induced corrections. Results have shown that the correction of corneal refractive power can be controlled with resolution of approximately 0.75 diopter. Inflation tests were performed post-treatment accordingly to assess the viscoelastic response of crosslinked corneas. Rate dependent hysteresis curves obtained from the pressure-deformation response of treated corneas under physiological conditions showed statistically significant increase in stiffness in contrast to controls, in which no change has been observed.

Maintaining a normal intraocular pressure (IOP) is important to visual health. Elevated IOPs have been implicated in many diseases, such as glaucoma and uveitis. The effects of an elevated IOP on the delicate tissues of the optic nervehead and retina are well-studied, but there is a lack of information about the effects of high IOPs on the stiffness of the crystalline lens. Changes in lenticular biomechanical properties have been implicated in diseases such as presbyopia and cataract, therefore, measuring lenticular biomechanical properties is crucial to understanding the etiology and progression of the leading causes of vision impairment. Additionally, there has been even less research focused on the effects of storage media on lenticular stiffness. Previous studies have been focused on the “gold standard” of mechanical testing on excised lenses. However, mechanical testing is invasive and destructive, and removal of the lens from the eyeglobe does not allow for properly replicating the lens environment in the eye-globe. Thus, there is a need for noninvasive measurement techniques capable of performing in situ and in vivo elastographic measurements of the lens. Here, we artificially controlled the IOP of whole porcine eye-globes (N=3). Acoustic radiation force induced low amplitude displacements (<10 μm) at the apex of the lenses, which then propagated as an elastic wave. The elastic wave propagation was detected by a phase-sensitive optical coherence elastography (PhS-OCE) system. The results show that the stiffness of the lenses increased when IOP increased from 10 mmHg IOP to 40 mmHg. Additional OCE measurements were made on excised lenses stored in various media (PBS, DMEM, and M-199) at different pHs (4-7) and at different temperatures (4°C, 22°C, and 37°C). The results show that the stiffness of the lenses increased slightly when incubated at 4°C or 22°C, but decreased when the lenses were incubated at 37°C, while lenses incubated in M-199 showed more stability in their stiffness than lenses incubated in PBS and DMEM. Moreover, the lenses stored in M-199 at a pH of 7 showed a decrease in stiffness over 24 hours, while the more acidic M-199 media caused an increase in lenticular stiffness.

Human color vision is achieved by mixing neural signals from cone photoreceptors sensitive to long- (L), medium- (M), and short- (S) wavelength light. The spatial arrangement and proportion of these spectral types in the retina set fundamental limits on color perception, and abnormal or missing types lead to color vision deficiencies. In vivo mapping of the trichromatic cone mosaic provides the most direct and quantitative means to assess the role photoreceptors play in color vision, but current methods of in vivo imaging have important limitations that preclude their widespread use. In this study, we present a new method for classifying cones based on their unique phase response to flashes of quasi-monochromatic light. Our use of phase provides unprecedented efficiency (30 min of subject time/retinal location) and accuracy (<0.02% of uncertainty), thus making in vivo cone classification practical in a wide range of color vision applications. We used adaptive optics optical coherence tomography to resolve cone cells in 3D and customized post-processing algorithms to extract the phase signal of individual cones. We successfully characterized light-induced changes to the phase signature of cones under different illuminant spectra, established the relationship between this phase change and the three cone spectral types, and used this relationship to classify and map cones in two color normal subjects.

The ganglion cell (GC) is the primary cell type damaged by diseases of the optic nerve such as glaucoma. Assessment of individual glaucoma risk is limited by our inability to accurately measure GC degeneration and loss. Recently, adaptive optics optical coherence tomography (AO-OCT) has enabled visualization and quantification of individual GC layer (GCL) somas in normal, healthy subjects. Quantifying GC loss in glaucoma, however, requires longitudinal assessment of these cells, which is confounded by normal age-related loss of these same cells. The ability to distinguish between these two causes of cell death is therefore paramount for early detection of glaucoma. In this study, we assess the ability of our AO-OCT method to track individual GCL somas over a period of one year and of our post processing methods to reliably measure soma loss rates. In four normal subjects with no history of ocular disease, we measured a soma loss rate of 0.15±0.04 %/yr (average±SD). As expected, this rate is more consistent with loss due to normal aging (~0.5%/yr) than to glaucomatous progression (~4.6%/yr). Aside from these rare isolated losses, the GCL soma mosaic was highly stable over the one year interval examined. Our measurements of peak GCL soma density did not differ significantly from histology reported in the literature.

Adaptive optics (AO) enables imaging of cellular structures in the retina that are not visible with clinical imaging techniques, providing the potential for earlier detection of retinal disease and enhanced monitoring of progression. The aim of this work was to develop automated, quantitative techniques for characterizing the morphology of the posterior retina in AO-OCT B-scans. Images were obtained from a custom-built dual-modality AO-optical coherence tomographyscanning laser ophthalmoscopy (AO-OCT-SLO) imaging system. Automated segmentation and cone identification procedures were developed and applied to images from two groups: healthy controls and dry age-related macular degeneration (AMD) subjects. Results from the automated routines were compared to measurements made manually by an expert human reviewer, demonstrating good agreement. Results from the control subjects showed decreasing cone inner and outer segment lengths with increasing distance from the fovea. The cone outer segment tip (COST) layer had a greater variation in axial position compared to the inner segment/outer segment (IS/OS) junction. Results from the AMD group indicate that disruption in the COST layer over drusen occurs earlier and to a greater extent than the IS/OS junction, which may be useful in the detection of emerging drusen.

We propose a study to better understand the impact of dynamic ocular aberrations in the axial resolution of nonconfocal adaptive optics (AO) ophthalmoscopes via a simulation of the 3D PSF in the retina for various AO-loop rates. We then use Optical Incoherence Tomography (OIT), a method enabling the generation of tomographic retinal cross-sections in incoherent imaging systems, to evaluate the benefits of a fast AO-loop rate on axial resolution and consequently on AO-corrected retinal image quality. We used the PARIS AO flood- illumination ophthalmoscope (FIO) for this study, where retinal images from different focal planes at an AO-loop rate of 10 Hz and 50 Hz were acquired.

Choriocapillaris is a unique vascular plexus located posterior to the retinal pigment epithelium. In the recent years, there is an increasing interest to investigate choriocapillaris alteration and progression of eye diseases and aging, using the optical coherence tomography angiography (OCTA). However, standardized algorithm for analysis has not been developed. Herein, we present non-invasive, in-vivo, high-resolution images of the non-human primates’ choriocapillaris using OCTA. Images were acquired with a prototype swept-source OCTA (SS-OCTA) system with 100kHz A-scan/s rate, over regions of 3×3 mm2 and 12×12 mm2. The non-perfusion area, also called flow voids, were segmented with an intensity damped, illuminance-compensated algorithm. The optimized quantification of the choriocapillaris flow voids may have applications in a wide array of eye diseases including age-related macular degeneration (AMD) and visualization of choriocapillaris in animal models could aid future studies on choroid involvement in models of eye disease.

We exploit the intrinsic phase stability of akinetic swept source optical coherence tomography to demonstrate digital defocus correction in-vivo at a center wavelength of 1060nm. The high speed of 500kHz enables digital adaptive optics (AO) correction across a field of view of 1.8x1.5deg, currently limited by the employed galvo scanners. The source operates in a previously presented dual resolution mode OCT system (wide field >40deg, AO >3deg) with hardware based adaptive optics. The latter allows to efficiently combine hardware and digital AO, and to further optimize the AO imaging results. We demonstrate the digitally assisted AO performance for both structural imaging as well as for OCT angiography imaging across the full retina down to the choriocapillaris.

Optical coherence tomography angiography (OCTA) is an extension of OCTA that allows for non-invasive imaging of the retinal microvasculature. OCTA imaging of adult retinal diseases is area of active research in ophthalmology as OCTA can provide insight into the pathogenesis of many retinal diseases. Like these adult diseases, pediatric diseases such as retinopathy of prematurity (ROP) have a primarily vascular pathogenesis. However, table top OCTA systems require compliant, seated subjects and cannot be used on infants and young children. In this manuscript we describe the development of a non-contact handheld OCTA (HH-OCTA) probe for imaging of young children and infants in the operating room. The probe utilizes a novel, diverging light on the scanner optical design that provides improved performance over a traditional OCT scanner design. While most handheld OCT probes are designed to be held by the side of the case or by a handle, our operators tend to prefer to grip probes by the tip of the probe for supine imagine. The ergonomics of the HH-OCTA probe were designed to match this grip. The HH-OCTA probe used a 200 kHz OCT engine, has a motorized stage that provides +10 to -10 D refractive error correction, and weighs 700g. Initial OCTA imaging was performed in 9 children or infants during exam under anesthesia. The HH-OCTA images provide visualization of the retinal microvasculature in both normal and pathological eyes.

There is a current void in efficient, cell-specific, retinal drug delivery systems, thus developing a safe, effective, selective drug delivery system would open novel therapeutic avenues. We previously demonstrated that femtosecond (fs) laser irradiation can selectively transfect DNA plasmids into cultured cells in the presence of functionalised gold nanoparticles (AuNPs) (1). Here, we sought out to selectively optoporate retinal cells in vivo with functionalized AuNPs and a 800nm fs laser. The cell-surface Kv1.1 voltage-gated channel was chosen to target retinal ganglion cells (RGCs) in the rat retina. The eyes of anesthetized rats were placed in the beam path of an optical system consisting of a fs laser and an ophthalmoscope for fundus visualization. Following Kv1.1-AuNP and FITC-dextran intravitreal injection and incubation, irradiation resulted in FITC uptake by retinal cells. In addition, similar experiments with Cy3-siRNA clearly show that the technique can effectively deliver siRNA into RGCs. Importantly, neither AuNP intravitreal injection nor irradiation resulted in RGC death, as determined by RBPMS quantification 1 week following AuNP injection and/or irradiation. Since living biological tissues absorb energy very weakly at 800nm, this non-invasive tool may provide a safe, cost effective approach to selectively target retinal cells and limit complications associated with surgical interventions, and potential biological hazards associated with viral-based gene therapy. In addition, given the extensive use of lasers in ophthalmic practice, our proposed technology may be seamlessly inserted to current clinical setups. (1) E. Bergeron et al, Nanoscale, 7, 17836 (2015).

To restore vision in patients who lost photoreceptors due to retinal degeneration, we developed a photovoltaic subretinal prosthesis which converts light into pulsed electric current, stimulating the inner retinal neurons. Visual information is projected onto the retina by video goggles using pulsed near-infrared (880nm) light. This design avoids the use of bulky electronics and wiring, thereby greatly reducing the surgical complexity and allows scaling the implants to thousands of electrodes. We found that similarly to normal vision, retinal response to prosthetic stimulation exhibits flicker fusion at high frequencies (>20Hz), adaptation to static images, antagonistic center-surround receptive fields with non-linear summation of its subunits. In rats, photovoltaic arrays with 55um pixels provided grating visual acuity up to a pixel pitch, which corresponds to about 20/200 acuity in a human eye. In patients with geographic atrophy, implants with 100um pixels provided retinotopically correct pattern percepts with resolution matching the pixel size. With flat pixels of 40um and smaller, stimulation thresholds are becoming prohibitively high. To reduce the pixel size further, we developed a novel honeycomb configuration of the stimulating electrode array with vertical walls separating the active and return electrodes, designed to leverage retinal migration for reducing the subretinal stimulation threshold and electrical cross-talk between neighboring pixels. Scalability, ease of implantation, and high resolution of these arrays open the door to highly functional restoration of sight in retinal degeneration.

Optical coherence tomographic angiography (OCT-A) technologies have been primarily demonstrated on slit-lamp systems, which preclude imaging in infants, bedridden patients, or patients who are otherwise unable to be imaged upright. Current-generation OCT-A requires densely-sampled volumetric datasets for high vascular resolution imaging, but bulk motion artifacts, resulting from saccades or eye drifts, often distort anatomic features during long acquisitions. Here, we demonstrate handheld motion-artifact corrected OCT-A using spectrally encoded coherence tomography and reflectometry (SECTR). SECTR has advantageous over previously demonstrated handheld ophthalmic imagers by acquiring spatiotemporally co-registered, high-speed en face images of the retinal fundus using spectrally encoded reflectometry (SER) concurrently with OCT. The orthogonal priority acquisition axes of SER and OCT enables volumetric registration and motion-artifact compensation. We have incorporated several optomechanical improvements including novel snap-fit lens mounts for reduced size and weight and improved optical stability over our previous design. Additionally, we developed a method for reducing back reflections from a double-clad fiber by fusion-splicing a no-core fiber segment with a predefined geometry. Lastly, we demonstrate in vivo human OCT-A imaging of the optic nerve head and fovea. OCT and OCT-A images were motion-corrected using complementary motion information extracted from en face SER and cross-sectional OCT images. Here, OCT-A volumetric datasets were densely-sampled in small regions-of-interest within a large SER field-of-view to achieve high vascular resolution OCT-A while maintaining sufficient fiducials within SER images for motion registration. We believe our probe will enable point-of-care functional ophthalmic imaging.

Monitoring retinal vascularization is crucial to understand the pathophysiology of major diseases affecting the retina. Laser Doppler holography addresses the problem of temporal resolution in blood flow imaging. The method is conceptually close to laser Doppler flowmetry except it uses digital holography which allows for full-field imaging with a simple Mach–Zehnder interferometer. The light backscattered by the retina is combined with a reference field in order to measure the beat frequency spectrum with a very high acquisition frame rate. Wideband measurements of the optical Doppler broadening were performed with a 75 kHz crash test camera. The power spectrum density of the recorded holograms was analyzed with a short-time Fourier transform analysis to reveal local pulsatile flow. By using laser Doppler holography, we were able to image blood flow in vivo and qualitatively in retinal vessels with a temporal resolution down to a few milliseconds which enabled to investigate blood flow profiles in arteries and veins. Additionally the angiographic contrast in power Doppler images has proved sensitive to lateral flow which made possible to image vessels in en-face planes. Finally we showed that laser Doppler holography allows to reveal non-invasively in young and healthy subjects the large vessels of the choroid. To this end we stitched together multiple power Doppler images to form a wide-field laser Doppler holographic panorama.

The inner plexiform layer (IPL) of the retina comprises extremely thin sublaminae with connections between bipolar cells, amacrine cells, and ganglion cells. So far, observations of IPL lamination in near-infrared Optical Coherence Tomography (OCT) images have been anecdotal. Visible light OCT theoretically provides higher axial resolution than near-infrared OCT for a given wavelength bandwidth. Imaging of the human retina with ultrahigh resolution visible light OCT and longitudinal chromatic aberration correction was recently shown, with a focus on the outer retina. Here, we demonstrate in vivo imaging of lamination in the inner plexiform layer using achromatized visible light Optical Coherence Tomography (OCT). To further improve the achievable axial resolution and contrast, we incorporate a grating light valve spatial light modulator (GLV-SLM) spectral shaping stage into our setup. The GLV-SLM rapidly and dynamically shapes the source spectrum to either reduce sidelobes in the axial point spread function, improve axial resolution by reducing the width of the axial point spread function, or switch between red light alignment mode and white light acquisition mode. In vivo retinal OCT images acquired from human subjects show that the IPL consists of 3 hyper-reflective bands and 2 hypo-reflective bands, corresponding well with the standard anatomical division of the IPL into 5 layers. Strategies to improve contrast of the subtle bands representing the IPL sublaminae are investigated. Possible explanations for the ability of visible light OCT to visualize IPL sublaminae, based only on backscattering or backreflection contrast, and implications for glaucoma progression monitoring, are discussed.

Research in neurobiology has identified a new ocular photoreceptor (melanopsin or ipRGC) which mediates a variety of light-based, non-visual effects on human physiology. One way to isolate the stimulation of ipRGCs is the silent substitution technique. We have built a Maxwellian view device capable of 85% ipRGCs contrast excitation with a large FOV (52^o). Four modulated LED light sources, illuminate a diffusing sphere, which exit aperture is imaged into the pupil of the eye. A camera with a 900 nm illumination capture the pupil. Without luminance changes (510±2 lm/m²), we increased ipRGC excitation from low to high level on three subjects. We observed a pupil constriction increasing with the ipRGC contrast. This suggests that we excite melanopsin silently. However, further experiments with electrophysiological and pupil recording needs to be done to completely validate our silent substitution device.

Purpose: Retinal diseases are the major cause of blindness in industrialized countries. A forecast reported that an estimated number of 196 million people will be affected by age related macular degeneration by 2020. While tremendous effort is made to develop novel therapeutic strategies to rescue retinal neurons and retinal pigment epithelium (RPE), optimal means to evaluate the effects of such treatments and diagnose the disease are still missing. Methods: We developed an imaging modality, called transscleral optical phase imaging (TOPI), which is able to resolve the individual human RPE cells in-vivo with the help of adaptive optics. The technology is based on oblique flood illumination and provides cellular resolution. The resulting 16 Hz-imaging speed, 5.7° × 5.7° field of view system allows for the visualization and the quantification of RPE cells within 2 seconds. Thanks to the approval from the ethic committee (CER-VD N°2017-00976), we conducted a study on 7 healthy human participants, with different skin pigmentations, 3 men and 4 women having an average age of 26 years. In all subjects, the RPE cell layer could be imaged and cell density could be quantified. Results: We show the RPE density and area analysis for 7 healthy subjects. The results of the analyses show comparable values to those found in the literature. Conclusion: The results of the study on healthy subjects demonstrate that TOPI is able to image and quantify in-vivo the human RPE cells, within a time frame of a few seconds (typically 2 seconds). The next step is to transfer the technique into a clinical environment.

Widefield ocular fundus imaging is conventionally performed in a reflection geometry. In this configuration, back-reflections from inner retinal layers, such as the nerve fiber layer, the inner limiting membrane, or even the anterior walls of large blood vessels, are often encountered, and may obscure the visibility of deeper features. Moreover, spectroscopic quantification of endogenous chromophores is complicated since the final image is a summation of reflections from several fundus layers (i.e. no single absorption pathlength can safely be assumed). Researchers have sought to model the fundus reflections, however the models are sensitive to the populations used and particular imaging platform. In theory, unwanted superficial reflections could be avoided and light path modeling could be simplified by adopting a transmission imaging geometry. We present an alternative transillumination fundus imaging strategy based on deeply penetrating near-infrared (NIR) light delivered transcranial near the subject’s temple. A portion of this light diffuses through bone and illuminates the posterior eye not from the front, as with conventional methods, but rather mostly from behind. As such, we image light transmitted through the fundus rather than back-reflected off multiple fundus layers. This single-pass measurement geometry simplifies absorption pathlength considerations and provides complementary information to fundus reflectometry. The use of NIR light enables imaging as deep as the choroid. Importantly, the technique is compatible with reflection-based techniques and we have shown that it works well with a commercial non-mydriatic fundus camera. Combining information from these two illumination approaches should improve spectroscopic analysis of the fundus.

Purpose: To identify early retinal biomarkers for Alzheimer disease (AD) using multimodal imaging. Methods: Infra-red (IR) and multicolor fundus imaging and spectral domain optic coherence tomography (SD-OCT) were performed in 108 offspring of AD patients (FH+) and 44 age-matched controls (FH-). All subjects were tested for cognitive function by executive function and episodic memory tests. MRI brain imaging was performed on a 3T MRI. Results: In FH+ subjects, lower performance in memory was associated with thicker peri-papillary temporal-superior RNFL (r=-0.220; p=.016). In FH- subjects, the correlation was in the opposite direction (r=0.335; p=.013). In FH+, left Hippocampal volume was associated with larger total macular thickness (r=0.212; p=.028), as well as thicker macular RNFL (r=0.216; p=.025), macular GCL (r=0.221; p=.022), and macular IPL (r=0.285, p=.003). Similar results were found in the right eye. Conclusions: The thickness of inner retinal layers and peripapillary RNFL are associated with cognitive functioning and hippocampal volume in asymptomatic subjects at high risk for AD and may present novel biomarkers for very early detection of AD.

Jones matrix optical coherence tomography (JM-OCT) is a functional extension of OCT. However, the clinical utility of JM-OCT is not widely accepted. Because of its hardware complexity and poorly established methods for clinical interpretation. In this study, we propose the approaches to solve the above-mentioned problems. To reduce the hardware complexity, we employ encapsulated passive polarization delay module (PPD) and encapsulated polarization diversity detection module (PDD), and develop full-function JM-OCT and simplified JM-OCT. In addition, we developed a pixel wise segmentation method for JM-OCT. The full-function JM-OCT which uses both PDD and PPD measures OCT, OCT angiography (OCTA), degree-of-polarization-uniformity (DOPU) and birefringence. The simplified JM-OCT which uses only PDD measures OCT, OCTA, and DOPU but not birefringence. In both JM-OCT systems, all the optical components are packed in a standard-sized retinal scanner. A pixel-wise segmentation method for retinal pigment epithelium (RPE) and choroidal stroma exploits multiple types of images obtained by the JM-OCT. Attenuation coefficient, OCTA, and DOPU are combined to synthesize a new artificial contrast. By applying a simple threshold to it, the target tissue is segmented. After segmenting the RPE, an en face “melano-layer thickness map” is created. A Normal subject and a pigment epithelial detachment (PED) subject are obtained by full-function JM-OCT and simplified JM-OCT. In PED subject, thickened RPE, hyper-reflective foci, and damaged RPE are correctly detected by RPE segmentation. In addition, created melano-layer thickness map has similar patterns to infrared fundus autofluorescence (NIR-AF), and it can contribute further interpretation of the NIR-AF.

Transparency of ocular structures is an important factor determining contrast in the retinal image. Although opacities are most commonly formed in the crystalline lens of aging eye (cataract formation), visual function can be also altered by the opacities in the vitreous body. Therefore, macro- and micro-scale visualization of vitreous is clinically relevant since alterations of vitreous organization impact retinal diseases and affect vision. However, optical imaging of the vitreous body is challenging due to its transparency. We demonstrate visualization of vitreous and its opacities in vivo using optical coherence tomography (OCT). We developed a prototype long-depth-range Swept-Source OCT instrument operating at the speed of 30 kA-scans/second and at the central wavelength of 1 μm to perform high-resolution imaging through the entire vitreous depth. The interface with focus-tunable optics has been used to optimize the field of view. 2-D and 3-D OCT data sets of eyes with vitreous opacities were acquired and processed to obtain contrast-enhanced high-resolution images of vitreous. The results demonstrate the ability of the OCT imaging to characterize the opacities that cause floaters. In conclusion, long-depth-range SS-OCT enables volumetric visualization of in vivo microstructural changes in the vitreous body. This instrument might be a useful tool in high-resolution evaluation and surgical management of vitreous opacities.

The present study aimed to develop a strategy for evaluation of instant PIMD-2π measurements as a basis for clinical monitoring of glaucoma. PIMD-2π is a morphometric measure of the waist of the nerve fiber layer at the optic nerve head (ONH). Clinical measurements of PIMD-2π in patients with early to moderate stage glaucoma demonstrated a high variability among subjects. The high variability among subjects renders comparison of instant PIMD-2π measurements to tolerance limits for normality derived from a normative database inefficient. It is suggested to instead compare sequential measurements of PIMD-2π within a patient. Initially, the difference between an instant measurement and the average of previous measurements can be compared to tolerance limits for difference between measurements within subject. Once, a potential loss of PIMD-2π is detected, a sufficient number of measurements within a sufficiently wide time interval can be used to estimate the PIMD-2π loss rate with regression and the deviation of the estimated loss rate can be evaluated as a 95 % confidence interval for the loss rate. If the upper confidence limit excludes 0, a significant loss rate has been detected. The currently proposed strategy has the potential to detect glaucoma earlier than the current gold standard, computer perimetry, with less inconvenience for the patient.

In this study, the human lacrimal canaliculus (LC), an important segment of the lacrimal drainage system, was imaged using dynamic optical coherence tomography (D-OCT) and the 3D structure of the LC was rendered. In D-OCT system, a turbid commercial ophthalmic solution was used as an extrinsic contrast agent. The lumen boundary of the LC appeared clearer in D-OCT images compared with simple static OCT images, making segmentation easier. D-OCT was performed by calculating the sum of the squared differences of intensities with two different normalization parameters. By color-combining these two D-OCT images and static OCT image, using image calculation software, the contrast agent and the lumen boundary can be clearly separated. 3D volumetric images of the LCs are demonstrated.

We report that a large field of view (FOV) retinal image can be acquired by a low-cost smart fundus camera. This handheld system includes a Raspberry Pi board, a touch screen display, a customized optical lens group, a ring LED light, and a Li-battery. Based on the open hardware platform and Linux operating system, we have integrated the portable system with component switching and image processing functions. Wide FOV has been realized using image stitching and structure from motion algorithms. This customized low-cost handheld fundus camera provides better image quality than cellphone-based fundus imaging solutions and offers more operational features than traditional portable fundus cameras. It may benefit field portable ophthalmic diagnostic applications.

The comprehensive architectural analysis of the retinal vasculature would greatly aid with the diagnosis and management of many ocular diseases. Optical coherence tomography angiography (OCTA) is a powerful micrometerlevel resolution, high sensitivity, and potentially large field of view retinal imaging modality that allows assessment of 2D and 3D microvascular networks. However, the quality of retinal OCTA images are often degraded by the noise and poor vascular contrast. Digital image filters are widely used in medical diagnostic to selectively enhance specific local intensity profiles or structures such as vasculatures. Most successful feature enhancement filters employ Hessian matrix and eigen values-based approach. In this paper, we demonstrate the feasibility of multiscale Hessian filters for the enhancement of the OCTA images of a mouse retina. We show that the enhancement filter based on the ratio of Hessian eigenvalues proposed by Jerman et al. (Jerman filter) performs better than the most commonly used Frangi’s method. This improved performance included close-to-uniform response in all vascular structures and enhancement of visibility of vascular structures with non-circular cross-sections. To evaluate and compare performance, different multi-scale Hessian filtering was performed on OCTA images of different inner retina vascular beds of a mouse eye.

Differential artery-vein analysis is valuable for early detection of diabetic retinopathy (DR) and other eye diseases. As a new optical coherence tomography (OCT) imaging modality, emerging OCT angiography (OCTA) provides capillary level resolution for accurate examination of retinal vasculatures. However, differential artery-vein analysis in OCTA, particularly for macular region in which blood vessels are small, is challenging. In coordination with an automatic vessel tracking algorithm, we report here the feasibility of using near infrared OCT oximetry to guide artery-vein classification in OCTA of macular region.

The hyaloid vascular system (HVS) is a transient capillary network nourishing developing eye. Currently, there is a lack of noninvasive imaging techniques for functional investigation of the HVS with high spatial and temporal resolutions. In this study, we demonstrated the feasibility of longitudinal optical coherence tomography angiography (OCTA) observation of the HVS regression in C57BL/6J mice. Longitudinal 3D OCT/OCTA measurements were conducted at 2- week, 3-week, and 4-week age by using a custom-designed OCT system. Three-dimensional OCT volume was acquired over 1.2 mm x 1.2 mm x 1.05 mm, containing a part of the lens tip, vitreous chamber, and the retina. Both OCT and OCTA successfully detected the hyaloid vessels, but OCTA benefits visualizing functional preservation of the blood vessels. OCTA images at 4-week age revealed functional loss of hyaloid vessels; while the OCT images still showed vascular remnants. We anticipate that longitudinal OCT/OCTA observation will be helpful to unravel the complex mechanism of the involution of the HVS correlated with eye development.

The cornea is a transparent tissue with significant refractive and barrier functions. The adult cornea has five layers: an outer epithelium layer, a Bowman’s layer which forms the transition between the corneal epithelium and the underlying stroma, a middle stromal layer of collagen-rich extracellular matrix between stromal keratocytes, a Descemet's membrane which separates the stroma from the underlying endothelial layer of the cornea and an inner layer of endothelial cells. Early studies showed that the that collagen fibrils of chick cornea display orthogonal-like pattern. Moreover, primary and secondary stroma have been identified in developing cornea. The primary stroma of the developing avian cornea is a highly organized extracellular matrix composed largely of striated collagen fibrils synthesized by the epithelium. Previous studies have found at least two different fibrillar collagen types, such as type I and II, are shown during the development. Other type of collagen, such as type IX, also appears to be involved in the collagen assembly process. This matrix is subsequently invaded by periocular mesenchymal cells which secrete a secondary cornea stroma to gradually replace the primary one. This indicates that the development of embryonic cornea involves many important biological events and exhibits highly dynamic characteristics. Since second-harmonic generation (SHG) is a nonlinear second order optical process which occurs in noncentrosymmetric systems with a large hyperpolarizability, it has emerged as a powerful modality for imaging fibrillar collagen in a diverse range of tissues. It is highly sensitive to the collagen fibril/fiber structure, and to changes that occur in diseases such as cancer, fibrosis and connective tissue disorders. To explore the structural variation of embryonic cornea, we use Fast Fourier Transform second harmonic generation microscopy as a tool in our study.

In clinical practice, ectatic disorders, such as keratoconus, are treated by accelerated corneal collagen crosslinking (ACXL). The treatment is based on the photodynamic reaction of riboflavin with ultraviolet A (UVA) light and increases the cornea’s mechanical stability. The clinical outcome of ACXL is usually evaluated several weeks post-treatment. An earlier evaluation could lead to a faster re-intervention in case of failure which could avoid additional discomfort and pain for the patient. We propose multiphoton tomography (MPT) to evaluate the outcome of ACXL soon after treatment.

In this study, we investigate ACXL-induced changes to the cornea autofluorescence (AF) using MPT. ACXL was performed in de-epithelialized corneal donor buttons and keratoconus corneas by infusing the samples with 0.1% riboflavin solution followed by UVA irradiation using either an in-house adapted system or a commercial ACXL system. AF lifetime images of the tissue were acquired prior and after treatment using MPT. As a control, corneas without treatment were monitored at the same time points.

Higher AF lifetimes were observed in the stroma of treated corneas than in control samples. The stroma AF lifetime was higher anteriorly, corresponding to the area where ACXL was most effective. First changes were observed as soon as 2 ℎ after treatment. We demonstrate that MPT can be used to follow-up the outcome of ACXL and that ACXL-induced changes can be detected sooner than with conventional methods and non-invasively.

Hyperspectral images of the human iris, containing spectral and spatial information of the iris, are acquired by a custom developed instrument. We developed this instrument specifically to perform in-vivo spectral reflectance measurements of the human iris. The significance of the data cube acquired is strictly related to the acquisition system performance. This paper analyzes the qualitative and quantitative in-vivo performances of our instrument considering in-vivo repeatability tests and comparing left and right irises of normal subjects, without visually assessed heterochromia of the irises.

Initial study found that depth-resolved refractive index variations encoded in retinal optical coherence tomography (OCT) exhibits multifractality. Interestingly, automated segmented different layers of the retina exhibited different degree of multifractality in a human eye. In an advanced study, we have adopted nano sensitive study of spectral contents of OCT signals before retinal image construction of a human eye. We have identified and constructed most contributed submicron structural spatial period OCT images. We have quantified nano structural morphological alteration in human retinal layers as deformation progress. This innovative approach promises to develop nano sensitive diagnostic tool for early disease detection in the human eye.

The corneal sub-basal nerve plexus (SNP) is a network of thin, unmyelinated nerve fibers located between the basal epithelium and the Bowman’s membrane. Both corneal and systemic diseases such as keratoconus and diabetic can alter the nerve fiber density, thickness and tortuosity. Recent developments of cellular resolution OCT technology allowed for in-vivo visualization and mapping of the corneal SNP. We have developed a fully automated algorithm for segmentation of corneal nerves. The performance of the algorithm was tested on a series of enface UHR-OCT images acquired in-vivo from healthy human subjects. The proposed algorithm traces most of the sub-basal corneal nerves correctly. The achieved processing time and tracing quality are the major advantages of the proposed method. Results show the potential application of proposed method for nerve analysis and morphometric quantification of human sub-basal corneal nerves which is an important tool in corneal related diseases.

We propose a novel concept of presbyopia-correcting adaptive eyeglasses for people who do not accept current corrective solutions. Our eyeglasses provide clear vision at all distances automatically and feature enhanced field of vision and high optical quality. The adaptive technology relies on an original fluid-filled lens whose focusing power is set by a low-power microfluidic pump inserted in the eyeglass temples. We present our prototype and demonstrate the viability of our technology through its characterization.

In femtosecond laser assisted keratoplasty different surgical wound profiles can be performed, such as mushroom, zigzag, anvil, Christmas tree, etc. The cut shape is chosen by the surgeon on the basis of patient’s morphology and pathology and on the gained experience. This work aims to qualitatively evaluate the biomechanical load resistance of the different configurations that are currently used in penetrating keratoplasty (PK), in order to support the surgeon’s choice. A 2D and a 3D finite-element biomechanical model of the human cornea was developed and different geometric configurations for PK were designed. The internal pressure was raised until the wound misaligned; wound prolapse then occurred. As a result, we evidenced a different wound resistance to internal loads in the different laser trephined profiles. The anvil profile was more resistant to the increasing internal pressure than was the mushroom or the zig zag pattern. This result is in accordance with the clinical results observed in previously treated patients. The anvil profile enabled the apposition of a restricted number of sutures and early suture removal, thanks to its greater mechanical load resistance. These advantages can contribute to a faster visual recovery in patients undergoing penetrating keratoplasty.

The QuantifEye-Macular Pigment Reflectometer (MPR) is a near commercial prototype that objectively measures lutein, zeaxanthin and overall macular pigment optical density (MPOD). We sought to evaluate the repeatability, intra and inter- observer variability and the effect of pupillary dilation on measurements obtained using the MPR and compare it with the subjective QuantifEye-MPS II, (MPSII). Experiment -1, thirty individuals were examined twice by a clinician. Measurements were performed for 40 seconds each eye and this dataset was parsed into various time intervals example 10-20, 10-30, 10-40 seconds etc. to evaluate the measurements’ repeatability and create clinical protocols that were utilized in the second experiment. Experiment-2 consists of two non-clinicians performing measurements on fifty individuals. Observer-1 performed measurements twice in the selected eye and once in the fellow eye both without and with pupillary dilation. The Observer-2 performed measurement only on the fellow eye both without and with pupillary dilation. Overall the MPR provides MPOD values that are well correlated with the MPSII. Of the various parsing of the data, the data 10-30 interval was the best at obtaining the MPOD, lutein and zeaxanthin values. The lutein and zeaxanthin optical densities were in the ranges of previously published histology results. Dilation was not needed to obtain the MPOD values but provided better lutein and zeaxanthin measurement repeatability. The intra and inter-observer repeatability in experiment-2 was excellent. The current studies have established the clinical protocols to use when measuring MPOD using the MPR and the MPR can provide repeatable and reliable measurements.

Adequate image quality is a necessary component to any retinal screening program whether the cases are to be read by a human reader or processed by an artificial intelligence system (AI). The need for expanded screening for retinal diseases has led to the adoption of low-cost, portable cameras that are ideal for reaching large underserved populations. However, the low-cost cameras generally require a higher level of operator skill to produce high quality images in comparison with more expensive table-top retinal cameras. This study, conducted at thirteen clinics in Monterrey, Mexico, compares unreadable rates between a table-top retinal camera (Canon CR2-AF) and a low-cost portable camera (Volk Pictor Plus) before and after implementation of automatic image quality assessment software, Image Quality Analyzer (IQA). The software determines if an image is of adequate quality to be read by a human or AI system; what the image quality issues are; and tips for fixing the issues. The process is performed in real time. Results show a significant decrease in unreadable cases (9% to 0%) for the Pictor Plus after IQA implementation bringing the percent of rejected cases in line with the table-top camera (3% to 5%).

Accurate refractive index of the contact lenses is one of the most important parameters needed to properly evaluate the lens performance. Measurements of the refractive power and other optical properties of the lenses are typically conducted in solution and converted to in-air performance. Such an approach is very sensitive to errors in the refractive indices of both the lens and the solution.

A number of different approaches have been undertaken by the manufacturers to overcome this difficulty. Efforts to accurately measure the phase refractive index a contact lens directly have for the most part been unsuccessful due to a curved shape of the lens and lens material properties. In this paper we describe a different approach of obtaining a group index dispersion curve with high accuracy of 0.003 or better, and discuss practical implications.

We have conducted measurements of the group refractive index (GRI) of hydrogel contact lenses at different wavelengths, from 530 to 670 nm, using time-domain low-coherence interferometer (LCI), based on the super-continuum light source (SCLS). The technique is being developed to overcome sources of error in the current measurement technique.