3DTINC: Time-Equivariant Non-Contrastive Learning for Predicting Disease Progression from Longitudinal OCTs.

Self-supervised learning (SSL) has emerged as a powerful technique for improving the efficiency and effectiveness of deep learning models. Contrastive methods are a prominent family of SSL that extract similar representations of two augmented views of an image while pushing away others in the representation space as negatives. However, the state-of-the-art contrastive methods require large batch sizes and augmentations designed for natural images that are impractical for 3D medical images. To address these limitations, we propose a new longitudinal SSL method, 3DTINC, based on non-contrastive learning. It is designed to learn perturbation-invariant features for 3D optical coherence tomography (OCT) volumes, using augmentations specifically designed for OCT. We introduce a new non-contrastive similarity loss term that learns temporal information implicitly from intra-patient scans acquired at different times. Our experiments show that this temporal information is crucial for predicting progression of retinal diseases, such as age-related macular degeneration (AMD). After pretraining with 3DTINC, we evaluated the learned representations and the prognostic models on two large-scale longitudinal datasets of retinal OCTs where we predict the conversion to wet-AMD within a six-month interval. Our results demonstrate that each component of our contributions is crucial for learning meaningful representations useful in predicting disease progression from longitudinal volumetric scans.

Morph-SSL: Self-Supervision with Longitudinal Morphing for Forecasting AMD Progression from OCT Volumes.

The lack of reliable biomarkers makes predicting the conversion from intermediate to neovascular age-related macular degeneration (iAMD, nAMD) a challenging task. We develop a Deep Learning (DL) model to predict the future risk of conversion of an eye from iAMD to nAMD from its current OCT scan. Although eye clinics generate vast amounts of longitudinal OCT scans to monitor AMD progression, only a small subset can be manually labeled for supervised DL. To address this issue, we propose Morph-SSL, a novel Self-supervised Learning (SSL) method for longitudinal data. It uses pairs of unlabelled OCT scans from different visits and involves morphing the scan from the previous visit to the next. The Decoder predicts the transformation for morphing and ensures a smooth feature manifold that can generate intermediate scans between visits through linear interpolation. Next, the Morph-SSL trained features are input to a Classifier which is trained in a supervised manner to model the cumulative probability distribution of the time to conversion with a sigmoidal function. Morph-SSL was trained on unlabelled scans of 399 eyes (3570 visits). The Classifier was evaluated with a five-fold cross-validation on 2418 scans from 343 eyes with clinical labels of the conversion date. The Morph-SSL features achieved an AUC of 0.779 in predicting the conversion to nAMD within the next 6 months, outperforming the same network when trained end-to-end from scratch or pre-trained with popular SSL methods. Automated prediction of the future risk of nAMD onset can enable timely treatment and individualized AMD management.

Detection of elusive DNA copy-number variations in hereditary disease and cancer through the use of noncoding and off-target sequencing reads.

Copy-number variants (CNVs) play a substantial role in the molecular pathogenesis of hereditary disease and cancer, as well as in normal human interindividual variation. However, they are still rather difficult to identify in mainstream sequencing projects, especially involving exome sequencing, because they often occur in DNA regions that are not targeted for analysis. To overcome this problem, we developed OFF-PEAK, a user-friendly CNV detection tool that builds on a denoising approach and the use of "off-target" DNA reads, which are usually discarded by sequencing pipelines. We benchmarked OFF-PEAK on data from targeted sequencing of 96 cancer samples, as well as 130 exomes of individuals with inherited retinal disease from three different populations. For both sets of data, OFF-PEAK demonstrated excellent performance (>95% sensitivity and >80% specificity vs. experimental validation) in detecting CNVs from in silico data alone, indicating its immediate applicability to molecular diagnosis and genetic research.

Retinal vessel volume reference database derived from volume-rendered optical coherence tomography angiography.

Optical coherence tomography angiography (OCTA) enables three-dimensional reconstruction of the functional blood vessels in the retina. Therefore, it enables the quantification of 3D retinal vessel parameters such as surface area and vessel volume. In spite of the widespread use of OCTA, no representative volume-rendered vessel volume (VV) data are published to date. In this study, OCTA 3 × 3 mm macular cubes were processed with volume-rendering techniques to measure VV in 203 eyes from 107 healthy volunteers. Generalized linear models (GLM) were constructed to assess the impact of age, gender, visual acuity (VA), spherical equivalent (SE), and axial length (AL) on VV. Overall mean VV was 0.23 ± 0.05mm3. Age and axial length showed a negative correlation with VV. However, GLM model analysis found that AL exerted the most pronounced influence on VV. No statistically significant associations were identified between gender or between left and right eyes. This is the first study to assess 3D OCTA VV and its naturally occurring variations in a large series of healthy subjects. It offers novel insights into the characterization of normal retinal vascular anatomy in healthy individuals, contributing to a valuable reference for future research in this field.

A SYSTEMATIC LITERATURE REVIEW OF DISEASE PROGRESSION REPORTED IN RPGR -ASSOCIATED X-LINKED RETINITIS PIGMENTOSA.

PURPOSE: Retinitis pigmentosa GTPase regulator-associated X-linked retinitis pigmentosa ( RPGR -associated XLRP) is a rare and severe form of retinitis pigmentosa, resulting in progressive visual impairment; however, disease progression data are limited. A systematic literature review was conducted to assess available data on disease progression in RPGR -associated XLRP. METHODS: PubMed, Embase, and select congress abstracts were evaluated through June 2022. Eligible studies included results specific to RPGR -associated XLRP or populations with ≥80% of patients with retinitis pigmentosa carrying disease-causing RPGR variants. End points of interest included visual acuity, visual field, ellipsoid zone width, progression to blindness, and patient-reported outcomes. RESULTS: Fourteen studies met ≥1 end point of interest. Progressive declines in visual acuity, visual field, and ellipsoid zone width were reported across studies. Nearly all publications reported annual declines in visual acuity (3.5%-8.2%). Annual visual field declines ranged from 4.2% to 13.3%. Changes in retinal structure were also observed (ellipsoid zone width changes: -177 to -830 µ m/year). Most studies measured blindness using visual acuity; visual field-based definitions resulted in blindness by age ∼25 years. Patient-reported outcome data were limited. CONCLUSION: Published evidence shows that patients with RPGR -associated XLRP experience progressive decline in visual acuity, visual field, and ellipsoid zone width, eventually resulting in blindness. Additional longitudinal data with standardized end points and expanded collection of patient-reported outcomes are needed to assess visual decline in RPGR -associated XLRP.

The biophysical and compositional properties of human basement membranes.

Basement membranes are among the most widespread, non-cellular functional materials in metazoan organisms. Despite this ubiquity, the links between their compositional and biophysical properties are often difficult to establish due to their thin and delicate nature. In this article, we examine these features on a molecular level by combining results from proteomics, elastic, and nanomechanical analyses across a selection of human basement membranes. Comparing results between these different membranes connects certain compositional attributes to distinct nanomechanical signatures and further demonstrates to what extent water defines these properties. In all, these data underline BMs as stiff yet highly elastic connective tissue layers and highlight how the interplay between composition, mechanics and hydration yields such exceptionally adaptable materials.

Progression of PROM1-Associated Retinal Degeneration as Determined by Spectral-Domain Optical Coherence Tomography Over a 24-Month Period.

PURPOSE: To evaluate the progression of atrophy as determined by spectral-domain optical coherence tomography (SD-OCT) in patients with molecularly confirmed PROM1-associated retinal degeneration (RD) over a 24-month period. DESIGN: International, multicenter, prospective case series. METHODS: A total of 13 eyes (13 patients) affected with PROM1-associated RD were enrolled at 5 sites and SD-OCT images were obtained at baseline and after 24 months. Loss of mean thickness (MT) and intact area were estimated after semi-automated segmentation for the following individual retinal layers in the central subfield (CS), inner ring, and outer ring of the ETDRS grid: retinal pigment epithelium (RPE), outer segments (OS), inner segments (IS), outer nuclear layer (ONL), inner retina (IR), and total retina (TR). RESULTS: Statistically significant losses of thickness of RPE and TR were detected in the CS and inner ring and of ONL and IS in the outer ring (all P < .05); a statistically significant decrease in the intact area of RPE and IS was observed in the inner ring, and of ONL in the outer ring (all P < .05); the change in MT and the intact area of the other layers showed a trend of decline over an observational period of 24 months. CONCLUSIONS: Significant thickness losses could be detected in outer retinal layers by SD-OCT over a 24-month period in patients with PROM1-associated retinal degeneration. Loss of thickness and/or intact area of such layers may serve as potential endpoints for clinical trials that aim to slow down the disease progression of PROM1-associated retinal degeneration.

Human selection bias drives the linear nature of the more ground truth effect in explainable deep learning optical coherence tomography image segmentation.

Supervised deep learning (DL) algorithms are highly dependent on training data for which human graders are assigned, for example, for optical coherence tomography (OCT) image annotation. Despite the tremendous success of DL, due to human judgment, these ground truth labels can be inaccurate and/or ambiguous and cause a human selection bias. We therefore investigated the impact of the size of the ground truth and variable numbers of graders on the predictive performance of the same DL architecture and repeated each experiment three times. The largest training dataset delivered a prediction performance close to that of human experts. All DL systems utilized were highly consistent. Nevertheless, the DL under-performers could not achieve any further autonomous improvement even after repeated training. Furthermore, a quantifiable linear relationship between ground truth ambiguity and the beneficial effect of having a larger amount of ground truth data was detected and marked as the more-ground-truth effect.

Establishing Fully-Automated Fundus-Controlled Dark Adaptometry: A Validation and Retest-Reliability Study.

Purpose: The purpose of this study was to establish and validate a novel fundus-controlled dark-adaptometry method. Methods: We developed a custom dark-adaptometry software for the S-MAIA device using the open-perimetry-interface. In the validation-substudy, participants underwent dark-adaptometry testing with a comparator device (MonCvONE, 59% rhodopsin bleach, cyan and red stimuli centered at 2 degrees, 4 degrees, and 6 degrees eccentricity). Following a brief break (approximately 5 minutes), the participants were bleached again and underwent dark-adaptometry testing with the S-MAIA device (same loci). In the retest reliability-substudy, participants were tested twice with the S-MAIA device (same loci as above). Nonlinear curve fitting was applied to extract dark-adaptation curve parameters. Validity and repeatability were summarized in terms of the mean bias and 95% limits of agreement (LoAs). Results: In the validation-substudy (N = 20 participants, median age interquartile range [IQR] 31.5 years [IQR = 25.8, 62.0]), measures of rod-mediated dark-adaptation showed little to no between method differences for the cone-rod-break-time (bias 95% confidence interval [95% CI] of +0.1 minutes [95% CI = -0.6 to 0.8]), rod-intercept-time (-0.23 minutes [95% CI = -1.38 to 0.93]), and S2 slope (-0.01 LogUnits/minutes [95% CI = -0.02 to -0.01]). In the retest reliability-substudy (N = 10 participants, 32.0 years [95% CI = 27.0, 57.5]), the corresponding LoAs were (cone-rod-break-time) -3.94 to 2.78 minutes, (rod-intercept-time) -4.55 to 3.11 minutes, and (S2 slope [rate-limited component of rod recovery]) -0.03 to 0.03 LogUnits/minutes. The LoAs for the steady-state cone and rod thresholds were -0.28 to 0.33 LogUnits and -0.34 to 0.28 LogUnits. Conclusions: The devised fundus-controlled dark-adaptometry method yields valid and reliable results. Translational Relevance: Fundus-controlled dark-adaptometry solves the critical need for localized testing of the visual cycle and retinoid transfer in eyes with unstable fixation.

Evaluating Contrast Sensitivity in Early and Intermediate Age-Related Macular Degeneration With the Quick Contrast Sensitivity Function.

Purpose: The purpose of this study was to describe, validate, and compare the contrast sensitivity functions (CSFs) acquired with the novel quick CSF (qCSF) method from patients with early and intermediate age-related macular degeneration (eAMD and iAMD) and healthy controls. Methods: This is a cross-sectional analysis of contrast sensitivity (CS) and visual acuity (VA) baseline data from the prospective Multimodal Functional and Structural Visual System Characterization (MUMOVI) study. The qCSF testing was conducted with the manifold contrast vision meter (Adaptive Sensory Technology, San Diego, CA, USA). CS levels at spatial frequencies from 1 cycle per degree (CPD) to 18 CPD, the area underneath the logarithmic contrast sensitivity function (AULCSF), and contrast acuity (CA) were analyzed. The association of functional metrics with variables of interest was tested with linear models. Results: Ninety-four study eyes from 94 study patients were included in the analysis (13 patients with eAMD, 33 patients with iAMD, and 48 healthy controls). Significant differences between the eAMD and the iAMD model estimates were only found for CS at 1 CPD (t value = -2.9, P value = 0.006) and CS at 1.5 CPD (-2.7, 0.01). A specific association between smoking years and CS at 1 CPD (P = 0.02) and CS at 1.5 CPD (P = 0.03) could be described in patients with AMD. Conclusions: The qCSF testing allows the fast measurement of the whole CSF, enabling the integration into clinical routine. We showed that novel qCSF-derived metrics detect slight functional differences between AMD stages, which testing by Pelli-Robson charts or VA testing would miss. This study, therefore, yields novel qCSF-derived candidate metrics for therapeutic trials in AMD.

Single versus Double PreserFlo MicroShunt Implantation in Glaucoma Patients: A Retrospective Cohort Study.

INTRODUCTION: The aim of this study was to describe and evaluate double PreserFlo MicroShunt implantation as a modified micro-invasive glaucoma surgery technique and to retrospectively compare the outcomes in a cohort of glaucoma patients with single or double implantation. MATERIALS AND METHODS: A retrospective data analysis of 57 glaucoma patients who consecutively underwent PreserFlo implantation was performed. Medical records were examined for patients' demographics, glaucoma type, intraocular pressure (IOP), medication, complications, and re-interventions. Two groups with single (n = 29) or double (n = 28) implantation were formed, and the outcomes were compared. In cases of two-stage double implantation (n = 17), the courses of the initial and the second implantations were compared. RESULTS: Mean preoperative IOP was significantly higher in the double compared to the single implantation group (29.4 ± 10.0 mm Hg; 21.7 ± 8.2 mm Hg; p = 0.003). Postoperatively, IOP was significantly lower in the double implantation group at various time-points (day 1, week 1, months 3 and 6; all p < 0.021). In the subgroup with two-stage procedures, mean preoperative IOP was 24.5 ± 8.5 mm Hg and 29.8 ± 10.1 mm Hg, respectively (p = 0.128). While immediately postoperatively, mean IOP lowering was clinically significant and similar following both procedures, the longer sustainable effect was observed after the second procedure (month 12: 25.5 ± 7.5 mm Hg; 12.4 ± 4.8 mm Hg; p = 0.001). No serious complications were observed. DISCUSSION/CONCLUSION: Double PreserFlo implantation appears safe and efficient for lowering IOP in glaucoma patients. Our preliminary findings suggest that double is superior to single implantation in terms of IOP lowering and the need for additional topical medication. Patients with insufficient IOP lowering following single implantation may benefit from a second implantation. Further research is warranted to evaluate double implantation as a first-line, one-stage procedure.

Automated deep learning-based AMD detection and staging in real-world OCT datasets (PINNACLE study report 5).

Real-world retinal optical coherence tomography (OCT) scans are available in abundance in primary and secondary eye care centres. They contain a wealth of information to be analyzed in retrospective studies. The associated electronic health records alone are often not enough to generate a high-quality dataset for clinical, statistical, and machine learning analysis. We have developed a deep learning-based age-related macular degeneration (AMD) stage classifier, to efficiently identify the first onset of early/intermediate (iAMD), atrophic (GA), and neovascular (nAMD) stage of AMD in retrospective data. We trained a two-stage convolutional neural network to classify macula-centered 3D volumes from Topcon OCT images into 4 classes: Normal, iAMD, GA and nAMD. In the first stage, a 2D ResNet50 is trained to identify the disease categories on the individual OCT B-scans while in the second stage, four smaller models (ResNets) use the concatenated B-scan-wise output from the first stage to classify the entire OCT volume. Classification uncertainty estimates are generated with Monte-Carlo dropout at inference time. The model was trained on a real-world OCT dataset, 3765 scans of 1849 eyes, and extensively evaluated, where it reached an average ROC-AUC of 0.94 in a real-world test set.

The Optical Coherence Tomography and Microperimetry Biomarker Evaluation in Patients with Geographic Atrophy (OMEGA) Study: Design and Baseline Characteristics - OMEGA Report 1.

INTRODUCTION: The aim of this study was to describe the design and the participants' baseline characteristics of a prospective natural history study of geographic atrophy (GA) secondary to age-related macular degeneration. METHODS: The optical coherence tomography (OCT) and microperimetry biomarker evaluation in patients with GA (OMEGA) study was conducted at a tertiary referral center (ClinicalTrials.gov identifier: NCT05963646). Participants were followed for 12 months during 4 visits (baseline and follow-up exams at weeks 12, 24, and 48) with best-corrected Early Treatment of Diabetic Retinopathy Study visual acuity, low-luminance visual acuity (LLVA), and quick contrast sensitivity function testing. Further, participants underwent spectral-domain OCT, OCT angiography, fundus autofluorescence imaging, and mesopic microperimetry testing. RESULTS: Thirty participants (median [IQR] age of 79 [77, 84] years) and 37 study eyes were included with a (median [IQR]) GA area of 1.40 mm2 (0.49, 5.24) at baseline. Out of 37 study eyes, six developed macular neovascularizations (16%). The study-eye best-corrected visual acuity was (median [IQR]) 0.18 logarithm of the minimum angle of resolution (logMAR) (0.06, 0.26), LLVA 0.66 logMAR (0.36, 0.88), and the microperimetry mean sensitivity 18.4 dB (9.21, 20.9). The highest correlation between square root GA area and a visual function test was evident for LLVA (R2 of 0.578), followed by area under the log contrast sensitivity function curve (0.519) and microperimetral retinal sensitivity (0.487). CONCLUSION: This report lays out the design and baseline characteristics of the OMEGA study, which aims to contribute to the understanding of the natural history of GA. The OMEGA study will provide estimates of the ability to detect change and retest reliability for a panel of structure and functional assessments.

Comparison of Novel Volumetric Microperimetry Metrics in Intermediate Age-Related Macular Degeneration: PINNACLE Study Report 3.

Purpose: To investigate and compare novel volumetric microperimetry (MP)-derived metrics in intermediate age-related macular degeneration (iAMD), as current MP metrics show high variability and low sensitivity. Methods: This is a cross-sectional analysis of microperimetry baseline data from the multicenter, prospective PINNACLE study (ClinicalTrials.gov NCT04269304). The Visual Field Modeling and Analysis (VFMA) software and an open-source implementation (OSI) were applied to calculate MP-derived hill-of-vison (HOV) surface plots and the total volume (VTOT) beneath the plots. Bland-Altman plots were used for methodologic comparison, and the association of retinal sensitivity metrics with explanatory variables was tested with mixed-effects models. Results: In total, 247 eyes of 189 participants (75 ± 7.3 years) were included in the analysis. The VTOT output of VFMA and OSI exhibited a significant difference (P < 0.0001). VFMA yielded slightly higher coefficients of determination than OSI and mean sensitivity (MS) in univariable and multivariable modeling, for example, in association with low-luminance visual acuity (LLVA) (marginal R2/conditional R2: VFMA 0.171/0.771, OSI 0.162/0.765, MS 0.133/0.755). In the multivariable analysis, LLVA was the only demonstrable predictor of VFMA VTOT (t-value, P-value: -7.5, <0.001) and MS (-6.5, <0.001). Conclusions: The HOV-derived metric of VTOT exhibits favorable characteristics compared to MS in evaluating retinal sensitivity. The output of VFMA and OSI is not exactly interchangeable in this cross-sectional analysis. Longitudinal analysis is necessary to assess their performance in ability-to-detect change. Translational Relevance: This study explores new volumetric MP endpoints for future application in therapeutic trials in iAMD and reports specific characteristics of the available HOV software applications.

Short-Term Effect of Micropulse Transscleral Laser Therapy on Intraocular Pressure in Untreated Fellow Eyes of Glaucoma Patients: Preliminary Results.

It has been observed that an intraocular pressure (IOP) altering intervention in one eye is followed by a consensual response in the untreated fellow eye. The underlying mechanisms remain unclear. Involvement of neuronal, cytokine, and hormonal regulation of aqueous humor dynamics, as well as improved treatment adherence or systemic absorption of topically administered medical compounds, have been suggested. Our aim was to investigate the short-term effects of unilateral micropulse transscleral laser therapy on IOP in the fellow eye. All medical records of glaucoma patients who underwent micropulse transscleral laser therapy in a tertiary referral center between May 2019 and February 2023 were collected and analyzed. We found a significant reduction in IOP in the treated eyes, indicating successful treatment. In the fellow eyes, despite not having changed any of the pharmacological IOP-reducing therapies, a significant reduction in IOP from 17.0 ± 5.1 mmHg to 13.5 ± 4.4 mmHg (p < 0.01) was observed. This reduction was, however, short-term and reached statistical significance on the first postoperative day only. Our findings support the concept of consensual inter-eye responses to unilateral IOP changes. Further research is warranted to elucidate the mechanisms underlying this phenomenon.

Validation of collaborative cyberspace virtual reality oculometry enhanced with near real-time spatial audio.

Currently, most medical image data, such as optical coherence tomography (OCT) images, are displayed in two dimensions on a computer screen. Advances in computer information technology have contributed to the growing storage of these data in electronic form. However, the data are usually processed only locally on site. To overcome such hurdles, a cyberspace virtual reality (csVR) application was validated, in which interactive OCT data were presented simultaneously to geographically distant sites (Lucerne, London, and Barcelona) where three graders independently measured the ocular csVR OCT diameters. A total of 109 objects were measured, each three times, resulting in a total of 327 csVR measurements. A minor mean absolute difference of 5.3 µm was found among the 3 measurements of an object (standard deviation 4.2 µm, coefficient of variation 0.3% with respect to the mean object size). Despite the 5 h of online work, csVR was well tolerated and safe. Digital high-resolution OCT data can be remotely and collaboratively processed in csVR. With csVR, measurements and actions enhanced with spatial audio communication can be made consistently in near real time, even if the users are situated geographically far apart. The proposed visuo-auditory framework has the potential to further boost the convenience of digital medicine toward csVR precision and collaborative medicine.

Multimodal spatiotemporal phenotyping of human retinal organoid development.

Organoids generated from human pluripotent stem cells provide experimental systems to study development and disease, but quantitative measurements across different spatial scales and molecular modalities are lacking. In this study, we generated multiplexed protein maps over a retinal organoid time course and primary adult human retinal tissue. We developed a toolkit to visualize progenitor and neuron location, the spatial arrangements of extracellular and subcellular components and global patterning in each organoid and primary tissue. In addition, we generated a single-cell transcriptome and chromatin accessibility timecourse dataset and inferred a gene regulatory network underlying organoid development. We integrated genomic data with spatially segmented nuclei into a multimodal atlas to explore organoid patterning and retinal ganglion cell (RGC) spatial neighborhoods, highlighting pathways involved in RGC cell death and showing that mosaic genetic perturbations in retinal organoids provide insight into cell fate regulation.

Exploring Healthy Retinal Aging with Deep Learning.

Purpose: To study the individual course of retinal changes caused by healthy aging using deep learning. Design: Retrospective analysis of a large data set of retinal OCT images. Participants: A total of 85 709 adults between the age of 40 and 75 years of whom OCT images were acquired in the scope of the UK Biobank population study. Methods: We created a counterfactual generative adversarial network (GAN), a type of neural network that learns from cross-sectional, retrospective data. It then synthesizes high-resolution counterfactual OCT images and longitudinal time series. These counterfactuals allow visualization and analysis of hypothetical scenarios in which certain characteristics of the imaged subject, such as age or sex, are altered, whereas other attributes, crucially the subject's identity and image acquisition settings, remain fixed. Main Outcome Measures: Using our counterfactual GAN, we investigated subject-specific changes in the retinal layer structure as a function of age and sex. In particular, we measured changes in the retinal nerve fiber layer (RNFL), combined ganglion cell layer plus inner plexiform layer (GCIPL), inner nuclear layer to the inner boundary of the retinal pigment epithelium (INL-RPE), and retinal pigment epithelium (RPE). Results: Our counterfactual GAN is able to smoothly visualize the individual course of retinal aging. Across all counterfactual images, the RNFL, GCIPL, INL-RPE, and RPE changed by -0.1 µm ± 0.1 µm, -0.5 µm ± 0.2 µm, -0.2 µm ± 0.1 µm, and 0.1 µm ± 0.1 µm, respectively, per decade of age. These results agree well with previous studies based on the same cohort from the UK Biobank population study. Beyond population-wide average measures, our counterfactual GAN allows us to explore whether the retinal layers of a given eye will increase in thickness, decrease in thickness, or stagnate as a subject ages. Conclusion: This study demonstrates how counterfactual GANs can aid research into retinal aging by generating high-resolution, high-fidelity OCT images, and longitudinal time series. Ultimately, we envision that they will enable clinical experts to derive and explore hypotheses for potential imaging biomarkers for healthy and pathologic aging that can be refined and tested in prospective clinical trials. Financial Disclosures: Proprietary or commercial disclosure may be found after the references.

Cynomolgus monkey's retina volume reference database based on hybrid deep learning optical coherence tomography segmentation.

Cynomolgus monkeys (Macaca fascicularis) are commonly used in pre-clinical ocular studies. However, studies that report the morphological features of the macaque retina are based only on minimal sample sizes; therefore, little is known about the normal distribution and background variation. This study was conducted using optical coherence tomography (OCT) imaging to investigate the variations in retinal volumes of healthy cynomolgus monkeys and the effects of sex, origin, and eye side on the retinal volumes to establish a comprehensive reference database. A machine-learning algorithm was employed to segment the retina within the OCT data (i.e., generated pixel-wise labels). Furthermore, a classical computer vision algorithm has identified the deepest point in a foveolar depression. The retinal volumes were determined and analyzed based on this reference point and segmented retinal compartments. Notably, the overall foveolar mean volume in zone 1, which is the region of the sharpest vision, was 0.205 mm3 (range 0.154-0.268 mm3), with a relatively low coefficient of variation of 7.9%. Generally, retinal volumes exhibit a relatively low degree of variation. However, significant differences in the retinal volumes due to the monkey's origin were identified. Additionally, sex had a significant impact on the paracentral retinal volume. Therefore, the origin and sex of cynomolgus monkeys should be considered when evaluating the macaque retinal volumes based on this dataset.