Quantitative Characterization of Retinal Features in Translated OCTA.

Purpose: This study explores the feasibility of using generative machine learning (ML) to translate Optical Coherence Tomography (OCT) images into Optical Coherence Tomography Angiography (OCTA) images, potentially bypassing the need for specialized OCTA hardware. Methods: The method involved implementing a generative adversarial network framework that includes a 2D vascular segmentation model and a 2D OCTA image translation model. The study utilizes a public dataset of 500 patients, divided into subsets based on resolution and disease status, to validate the quality of TR-OCTA images. The validation employs several quality and quantitative metrics to compare the translated images with ground truth OCTAs (GT-OCTA). We then quantitatively characterize vascular features generated in TR-OCTAs with GT-OCTAs to assess the feasibility of using TR-OCTA for objective disease diagnosis. Result: TR-OCTAs showed high image quality in both 3 and 6 mm datasets (high-resolution, moderate structural similarity and contrast quality compared to GT-OCTAs). There were slight discrepancies in vascular metrics, especially in diseased patients. Blood vessel features like tortuosity and vessel perimeter index showed a better trend compared to density features which are affected by local vascular distortions. Conclusion: This study presents a promising solution to the limitations of OCTA adoption in clinical practice by using vascular features from TR-OCTA for disease detection. Translation relevance: This study has the potential to significantly enhance the diagnostic process for retinal diseases by making detailed vascular imaging more widely available and reducing dependency on costly OCTA equipment.

TENAYA and LUCERNE: Two-Year Results from the Phase 3 Neovascular Age-Related Macular Degeneration Trials of Faricimab with Treat-and-Extend Dosing in Year 2.

PURPOSE: To evaluate 2-year efficacy, durability, and safety of the bispecific antibody faricimab, which inhibits both angiopoietin-2 and VEGF-A. DESIGN: TENAYA (ClinicalTrials.gov identifier, NCT03823287) and LUCERNE (ClinicalTrials.gov identifier, NCT03823300) were identically designed, randomized, double-masked, active comparator-controlled phase 3 noninferiority trials. PARTICIPANTS: Treatment-naive patients with neovascular age-related macular degeneration (nAMD) 50 years of age or older. METHODS: Patients were randomized (1:1) to intravitreal faricimab 6.0 mg up to every 16 weeks (Q16W) or aflibercept 2.0 mg every 8 weeks (Q8W). Faricimab fixed dosing based on protocol-defined disease activity at weeks 20 and 24 up to week 60, followed up to week 108 by a treat-and-extend personalized treatment interval regimen. MAIN OUTCOME MEASURES: Efficacy analyses included change in best-corrected visual acuity (BCVA) from baseline at 2 years (averaged over weeks 104, 108, and 112) and proportion of patients receiving Q16W, every 12 weeks (Q12W), and Q8W dosing at week 112 in the intention-to-treat population. Safety analyses included ocular adverse events (AEs) in the study eye through study end at week 112. RESULTS: Of 1326 patients treated across TENAYA/LUCERNE, 1113 (83.9%) completed treatment (n = 555 faricimab; n = 558 aflibercept). The BCVA change from baseline at 2 years was comparable between faricimab and aflibercept groups in TENAYA (adjusted mean change, +3.7 letters [95% confidence interval (CI), +2.1 to +5.4] and +3.3 letters [95% CI, +1.7 to +4.9], respectively; mean difference, +0.4 letters [95% CI, -1.9 to +2.8]) and LUCERNE (adjusted mean change, +5.0 letters [95% CI, +3.4 to +6.6] and +5.2 letters [95% CI, +3.6 to +6.8], respectively; mean difference, -0.2 letters [95% CI, -2.4 to +2.1]). At week 112 in TENAYA and LUCERNE, 59.0% and 66.9%, respectively, achieved Q16W faricimab dosing, increasing from year 1, and 74.1% and 81.2%, achieved Q12W or longer dosing. Ocular AEs in the study eye were comparable between faricimab and aflibercept groups in TENAYA (55.0% and 56.5% of patients, respectively) and LUCERNE (52.9% and 47.5% of patients, respectively) through week 112. CONCLUSIONS: Treat-and-extend faricimab treatment based on nAMD disease activity maintained vision gains through year 2, with most patients achieving extended dosing intervals. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

METformin for the MINimization of Geographic Atrophy Progression (METforMIN): A Randomized Trial.

Purpose: Metformin use has been associated with a decreased risk of age-related macular degeneration (AMD) progression in observational studies. We aimed to evaluate the efficacy of oral metformin for slowing geographic atrophy (GA) progression. Design: Parallel-group, multicenter, randomized phase II clinical trial. Participants: Participants aged ≥ 55 years without diabetes who had GA from atrophic AMD in ≥ 1 eye. Methods: We enrolled participants across 12 clinical centers and randomized participants in a 1:1 ratio to receive oral metformin (2000 mg daily) or observation for 18 months. Fundus autofluorescence imaging was obtained at baseline and every 6 months. Main Outcome Measures: The primary efficacy endpoint was the annualized enlargement rate of the square root-transformed GA area. Secondary endpoints included best-corrected visual acuity (BCVA) and low luminance visual acuity (LLVA) at each visit. Results: Of 66 enrolled participants, 34 (57 eyes) were randomized to the observation group and 32 (53 eyes) were randomized to the treatment group. The median follow-up duration was 13.9 and 12.6 months in the observation and metformin groups, respectively. The mean ± standard error annualized enlargement rate of square root transformed GA area was 0.35 ± 0.04 mm/year in the observation group and 0.42 ± 0.04 mm/year in the treatment group (risk difference = 0.07 mm/year, 95% confidence interval = -0.05 to 0.18 mm/year; P = 0.26). The mean ± standard error decline in BCVA was 4.8 ± 1.7 letters/year in the observation group and 3.4 ± 1.1 letters/year in the treatment group (P = 0.56). The mean ± standard error decline in LLVA was 7.3 ± 2.5 letters/year in the observation group and 0.8 ± 2.2 letters/year in the treatment group (P = 0.06). Fourteen participants in the metformin group experienced nonserious adverse events related to metformin, with gastrointestinal side effects as the most common. No serious adverse events were attributed to metformin. Conclusions: The results of this trial as conducted do not support oral metformin having effects on reducing the progression of GA. Additional placebo-controlled trials are needed to explore the role of metformin for AMD, especially for earlier stages of the disease. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Optimizing the OCTA layer fusion option for deep learning classification of diabetic retinopathy.

The purpose of this study is to evaluate layer fusion options for deep learning classification of optical coherence tomography (OCT) angiography (OCTA) images. A convolutional neural network (CNN) end-to-end classifier was utilized to classify OCTA images from healthy control subjects and diabetic patients with no retinopathy (NoDR) and non-proliferative diabetic retinopathy (NPDR). For each eye, three en-face OCTA images were acquired from the superficial capillary plexus (SCP), deep capillary plexus (DCP), and choriocapillaris (CC) layers. The performances of the CNN classifier with individual layer inputs and multi-layer fusion architectures, including early-fusion, intermediate-fusion, and late-fusion, were quantitatively compared. For individual layer inputs, the superficial OCTA was observed to have the best performance, with 87.25% accuracy, 78.26% sensitivity, and 90.10% specificity, to differentiate control, NoDR, and NPDR. For multi-layer fusion options, the best option is the intermediate-fusion architecture, which achieved 92.65% accuracy, 87.01% sensitivity, and 94.37% specificity. To interpret the deep learning performance, the Gradient-weighted Class Activation Mapping (Grad-CAM) was utilized to identify spatial characteristics for OCTA classification. Comparative analysis indicates that the layer data fusion options can affect the performance of deep learning classification, and the intermediate-fusion approach is optimal for OCTA classification of DR.

Federated learning for diagnosis of age-related macular degeneration.

This paper presents a federated learning (FL) approach to train deep learning models for classifying age-related macular degeneration (AMD) using optical coherence tomography image data. We employ the use of residual network and vision transformer encoders for the normal vs. AMD binary classification, integrating four unique domain adaptation techniques to address domain shift issues caused by heterogeneous data distribution in different institutions. Experimental results indicate that FL strategies can achieve competitive performance similar to centralized models even though each local model has access to a portion of the training data. Notably, the Adaptive Personalization FL strategy stood out in our FL evaluations, consistently delivering high performance across all tests due to its additional local model. Furthermore, the study provides valuable insights into the efficacy of simpler architectures in image classification tasks, particularly in scenarios where data privacy and decentralization are critical using both encoders. It suggests future exploration into deeper models and other FL strategies for a more nuanced understanding of these models' performance. Data and code are available at https://github.com/QIAIUNCC/FL_UNCC_QIAI.

Simulation of long-term impact of intravitreal anti-VEGF therapy on patients with severe non-proliferative diabetic retinopathy.

OBJECTIVE: A simulation model was constructed to assess long-term outcomes of proactively treating severe non-proliferative diabetic retinopathy (NPDR) with anti-vascular endothelial growth factor (anti-VEGF) therapy versus delaying treatment until PDR develops. METHODS AND ANALYSIS: Simulated patients were generated using a retrospective real-world cohort of treatment-naive patients identified in an electronic medical records database (IBM Explorys) between 2011 and 2017. Impact of anti-VEGF treatment was derived from clinical trial data for intravitreal aflibercept (PANORAMA) and ranibizumab (RISE/RIDE), averaged by weighted US market share. Real-world risk of PDR progression was modelled using Cox multivariable regression. The Monte Carlo simulation model examined rates of progression to PDR and sustained blindness (visual acuity <20/200) for 2 million patients scaled to US NPDR disease prevalence. Simulated progression rates from severe NPDR to PDR over 5 years and blindness rates over 10 years were compared for delayed versus early-treatment patients. RESULTS: Real-world data from 77 454 patients with mild-to-severe NPDR simulated 2 million NPDR patients, of which 86 680 had severe NPDR. Early treatment of severe NPDR with anti-VEGF therapy led to a 51.7% relative risk reduction in PDR events over 5 years (15 704 early vs 32 488 delayed), with a 19.4% absolute risk reduction (18.1% vs 37.5%). Sustained blindness rates at 10 years were 4.4% for delayed and 1.9% for early treatment of severe NPDR. CONCLUSION: The model suggests treating severe NPDR early with anti-VEGF therapy, rather than delaying treatment until PDR develops, could significantly reduce PDR incidence over 5 years and sustained blindness over 10 years.

An open-source deep learning network AVA-Net for arterial-venous area segmentation in optical coherence tomography angiography.

BACKGROUND: Differential artery-vein (AV) analysis in optical coherence tomography angiography (OCTA) holds promise for the early detection of eye diseases. However, currently available methods for AV analysis are limited for binary processing of retinal vasculature in OCTA, without quantitative information of vascular perfusion intensity. This study is to develop and validate a method for quantitative AV analysis of vascular perfusion intensity. METHOD: A deep learning network AVA-Net has been developed for automated AV area (AVA) segmentation in OCTA. Seven new OCTA features, including arterial area (AA), venous area (VA), AVA ratio (AVAR), total perfusion intensity density (T-PID), arterial PID (A-PID), venous PID (V-PID), and arterial-venous PID ratio (AV-PIDR), were extracted and tested for early detection of diabetic retinopathy (DR). Each of these seven features was evaluated for quantitative evaluation of OCTA images from healthy controls, diabetic patients without DR (NoDR), and mild DR. RESULTS: It was observed that the area features, i.e., AA, VA and AVAR, can reveal significant differences between the control and mild DR. Vascular perfusion parameters, including T-PID and A-PID, can differentiate mild DR from control group. AV-PIDR can disclose significant differences among all three groups, i.e., control, NoDR, and mild DR. According to Bonferroni correction, the combination of A-PID and AV-PIDR can reveal significant differences in all three groups. CONCLUSIONS: AVA-Net, which is available on GitHub for open access, enables quantitative AV analysis of AV area and vascular perfusion intensity. Comparative analysis revealed AV-PIDR as the most sensitive feature for OCTA detection of early DR. Ensemble AV feature analysis, e.g., the combination of A-PID and AV-PIDR, can further improve the performance for early DR assessment.

Some people with diabetes develop diabetic retinopathy, in which the blood flow through the eye changes, resulting in damage to the back of the eye, called the retina. Changes in blood flow can be measured by imaging the eye using a method called optical coherence tomography angiography (OCTA). The authors developed a computer program named AVA-Net that determines changes in blood flow through the eye from OCTA images. The program was tested on images from people with healthy eyes, people with diabetes but no eye disease, and people with mild diabetic retinopathy. Their program found differences between these groups and so could be used to improve diagnosis of people with diabetic retinopathy.

Contrastive learning-based pretraining improves representation and transferability of diabetic retinopathy classification models.

Diabetic retinopathy (DR) is a major cause of vision impairment in diabetic patients worldwide. Due to its prevalence, early clinical diagnosis is essential to improve treatment management of DR patients. Despite recent demonstration of successful machine learning (ML) models for automated DR detection, there is a significant clinical need for robust models that can be trained with smaller cohorts of dataset and still perform with high diagnostic accuracy in independent clinical datasets (i.e., high model generalizability). Towards this need, we have developed a self-supervised contrastive learning (CL) based pipeline for classification of referable vs non-referable DR. Self-supervised CL based pretraining allows enhanced data representation, therefore, the development of robust and generalized deep learning (DL) models, even with small, labeled datasets. We have integrated a neural style transfer (NST) augmentation in the CL pipeline to produce models with better representations and initializations for the detection of DR in color fundus images. We compare our CL pretrained model performance with two state of the art baseline models pretrained with Imagenet weights. We further investigate the model performance with reduced labeled training data (down to 10 percent) to test the robustness of the model when trained with small, labeled datasets. The model is trained and validated on the EyePACS dataset and tested independently on clinical datasets from the University of Illinois, Chicago (UIC). Compared to baseline models, our CL pretrained FundusNet model had higher area under the receiver operating characteristics (ROC) curve (AUC) (CI) values (0.91 (0.898 to 0.930) vs 0.80 (0.783 to 0.820) and 0.83 (0.801 to 0.853) on UIC data). At 10 percent labeled training data, the FundusNet AUC was 0.81 (0.78 to 0.84) vs 0.58 (0.56 to 0.64) and 0.63 (0.60 to 0.66) in baseline models, when tested on the UIC dataset. CL based pretraining with NST significantly improves DL classification performance, helps the model generalize well (transferable from EyePACS to UIC data), and allows training with small, annotated datasets, therefore reducing ground truth annotation burden of the clinicians.

Normalized Blood Flow Index in Optical Coherence Tomography Angiography Provides a Sensitive Biomarker of Early Diabetic Retinopathy.

Purpose: To evaluate the sensitivity of normalized blood flow index (NBFI) for detecting early diabetic retinopathy (DR). Methods: Optical coherence tomography angiography (OCTA) images of healthy controls, diabetic patients without DR (NoDR), and patients with mild nonproliferative DR (NPDR) were analyzed in this study. The OCTA images were centered on the fovea and covered a 6 mm × 6 mm area. Enface projections of the superficial vascular plexus (SVP) and the deep capillary plexus (DCP) were obtained for the quantitative OCTA feature analysis. Three quantitative OCTA features were examined: blood vessel density (BVD), blood flow flux (BFF), and NBFI. Each feature was calculated from both the SVP and DCP and their sensitivities to distinguish the three cohorts of the study were evaluated. Results: The only quantitative feature capable of distinguishing all three cohorts was NBFI in the DCP image. Comparative study revealed that both BVD and BFF were able to distinguish the controls and NoDR from mild NPDR. However, neither BVD nor BFF was sensitive enough to separate NoDR from the healthy controls. Conclusions: The NBFI has been demonstrated as a sensitive biomarker of early DR, revealing retinal blood flow abnormality better than traditional BVD and BFF. The NBFI in the DCP was verified as the most sensitive biomarker, supporting that diabetes affects the DCP earlier than SVP in DR. Translational Relevance: NBFI provides a robust biomarker for quantitative analysis of DR-caused blood flow abnormalities, promising early detection and objective classification of DR.

Visual Field Abnormalities in Early-Stage Diabetic Retinopathy Assessed by Chromatic Perimetry.

Purpose: The purpose of this study was to define the nature and extent of sensitivity loss using chromatic perimetry in diabetics who have mild or no retinopathy. Methods: Thirty-four individuals with type II diabetes mellitus who have mild nonproliferative diabetic retinopathy (MDR; N = 17) or no diabetic retinopathy (NDR; N = 17) and 15 visually normal, non-diabetic controls participated. Sensitivity was assessed along the horizontal visual field meridian using an Octopus 900 perimeter. Measurements were performed under light- and dark-adapted conditions using long-wavelength (red) and short-wavelength (blue) Goldmann III targets. Cumulative defect curves (CDCs) were constructed to determine whether field sensitivity loss was diffuse or localized. Results: Sensitivity was reduced significantly under light-adapted conditions for both stimulus colors for the NDR (mean defect ± SEM = -2.1 dB ± 0.6) and MDR (mean defect ± SEM = -4.0 dB ± 0.7) groups. Sensitivity was also reduced under dark-adapted conditions for both stimulus colors for the NDR (mean defect ± SEM = -1.9 dB ± 0.7) and MDR (mean defect ± SEM = -4.5 ± 1.0 dB) groups. For both diabetic groups, field loss tended to be diffuse under light-adapted conditions (up to 6.9 dB loss) and localized under dark-adapted conditions (up to 15.4 dB loss). Conclusions: Visual field sensitivity losses suggest neural abnormalities in early stage diabetic eye disease and the pattern of the sensitivity losses differed depending on the adaptation conditions. Chromatic perimetry may be useful for subtyping individuals who have mild or no diabetic retinopathy and for better understanding their neural dysfunction.

COMPARATIVE ANALYSIS OF OCT AND OCT ANGIOGRAPHY CHARACTERISTICS IN EARLY DIABETIC RETINOPATHY.

PURPOSE: To assess the quantitative characteristics of optical coherence tomography (OCT) and OCT angiography (OCTA) for the objective detection of early diabetic retinopathy (DR). METHODS: This was a retrospective and cross-sectional study, which was carried out at a tertiary academic practice with a subspecialty. Twenty control participants, 15 people with diabetics without retinopathy (NoDR), and 22 people with mild nonproliferative diabetic retinopathy (NPDR) were included in this study. Quantitative OCT characteristics were derived from the photoreceptor hyperreflective bands, i.e., inner segment ellipsoid (ISe) and retinal pigment epithelium (RPE). OCTA characteristics, including vessel diameter index (VDI), vessel perimeter index (VPI), and vessel skeleton density (VSD), were evaluated. RESULTS: Quantitative OCT analysis indicated that the ISe intensity was significantly trending downward with DR advancement. Comparative OCTA revealed VDI, VPI, and VSD as the most sensitive characteristics of DR. Correlation analysis of OCT and OCTA characteristics revealed weak variable correlation between the two imaging modalities. CONCLUSION: Quantitative OCT and OCTA analyses revealed photoreceptor and vascular distortions in early DR. Comparative analysis revealed that the OCT intensity ratio, ISe/RPE, has the best sensitivity for early DR detection. Weak variable correlation of the OCT and OCTA characteristics suggests that OCT and OCTA are providing supplementary information for DR detection and classification.

Rhegmatogenous retinal detachment surgery: A review.

Rhegmatogenous retinal detachment (RRD) is a serious surgical condition with significant ocular morbidity if not managed properly. Once untreatable, approaches to the repair of RRD have greatly evolved over the years, leading to outstanding primary surgical success rates. The management of RRD is often a topic of great debate. Scleral buckling, vitrectomy and pneumatic retinopexy have been used successfully for the treatment of RRD. Several factors may affect surgical success and dictate a surgeon's preference for the technique employed. In this review, we provide an overview and supporting literature on the options for RRD repair and their respective preoperative and postoperative considerations in order to guide surgical management.

Automated Region of Interest Selection Improves Deep Learning-Based Segmentation of Hyper-Reflective Foci in Optical Coherence Tomography Images.

Hyperreflective foci (HRF) have been associated with retinal disease progression and demonstrated as a negative prognostic biomarker for visual function. Automated segmentation of HRF in retinal optical coherence tomography (OCT) scans can be beneficial to identify the formation and movement of the HRF biomarker as a retinal disease progresses and can serve as the first step in understanding the nature and severity of the disease. In this paper, we propose a fully automated deep neural network based HRF segmentation model in OCT images. We enhance the model's performance by using a patch-based strategy that increases the model's compute on the HRF pixels. The patch-based strategy is evaluated against state of the art HRF segmentation pipelines on clinical retinal image data. Our results shows that the patch-based approach demonstrates a high precision score and intersection over union (IOU) using a ResNet34 segmentation model with Binary Cross Entropy loss function. The HRF segmentation pipeline can be used for analyzing HRF biomarkers for different retinopathies.

MF-AV-Net: an open-source deep learning network with multimodal fusion options for artery-vein segmentation in OCT angiography.

This study is to demonstrate the effect of multimodal fusion on the performance of deep learning artery-vein (AV) segmentation in optical coherence tomography (OCT) and OCT angiography (OCTA); and to explore OCT/OCTA characteristics used in the deep learning AV segmentation. We quantitatively evaluated multimodal architectures with early and late OCT-OCTA fusions, compared to the unimodal architectures with OCT-only and OCTA-only inputs. The OCTA-only architecture, early OCT-OCTA fusion architecture, and late OCT-OCTA fusion architecture yielded competitive performances. For the 6 mm×6 mm and 3 mm×3 mm datasets, the late fusion architecture achieved an overall accuracy of 96.02% and 94.00%, slightly better than the OCTA-only architecture which achieved an overall accuracy of 95.76% and 93.79%. 6 mm×6 mm OCTA images show AV information at pre-capillary level structure, while 3 mm×3 mm OCTA images reveal AV information at capillary level detail. In order to interpret the deep learning performance, saliency maps were produced to identify OCT/OCTA image characteristics for AV segmentation. Comparative OCT and OCTA saliency maps support the capillary-free zone as one of the possible features for AV segmentation in OCTA. The deep learning network MF-AV-Net used in this study is available on GitHub for open access.

QUANTITATIVE OPTICAL COHERENCE TOMOGRAPHY REVEALS ROD PHOTORECEPTOR DEGENERATION in EARLY DIABETIC RETINOPATHY.

PURPOSE: This study is to test the feasibility of optical coherence tomography (OCT) detection of photoreceptor abnormality and to verify that the photoreceptor abnormality is rod predominated in early diabetic retinopathy (DR). METHODS: OCT images were acquired from normal eyes, diabetic eyes with no DR, and mild nonproliferative DR (NPDR). Quantitative features, including thickness measurements quantifying band distances and reflectance intensity features among the external limiting membrane, inner segment ellipsoid, interdigitation zone, and retinal pigment epithelium were determined. Comparative OCT analysis of central fovea, parafovea, and perifovea were implemented to verify that the photoreceptor abnormality is rod predominated in early DR. RESULTS: Thickness abnormalities between the inner segment ellipsoid and interdigitation zone also showed a decreasing trend among cohorts. Reflectance abnormalities of the external limiting membrane, interdigitation zone, and inner segment ellipsoid were observed between healthy, no DR, and mild NPDR eyes. The normalized inner segment ellipsoid/retinal pigment epithelium intensity ratio revealed a significant decreasing trend in the perifovea, but no detectable difference in central fovea. CONCLUSION: Quantitative OCT analysis consistently revealed outer retina, i.e., photoreceptor changes in diabetic patients with no DR and mild NPDR. Comparative analysis of central fovea, parafovea, and perifovea confirmed that the photoreceptor abnormality is rod-predominated in early DR.

From Data to Deployment: The Collaborative Community on Ophthalmic Imaging Roadmap for Artificial Intelligence in Age-Related Macular Degeneration.

OBJECTIVE: Health care systems worldwide are challenged to provide adequate care for the 200 million individuals with age-related macular degeneration (AMD). Artificial intelligence (AI) has the potential to make a significant, positive impact on the diagnosis and management of patients with AMD; however, the development of effective AI devices for clinical care faces numerous considerations and challenges, a fact evidenced by a current absence of Food and Drug Administration (FDA)-approved AI devices for AMD. PURPOSE: To delineate the state of AI for AMD, including current data, standards, achievements, and challenges. METHODS: Members of the Collaborative Community on Ophthalmic Imaging Working Group for AI in AMD attended an inaugural meeting on September 7, 2020, to discuss the topic. Subsequently, they undertook a comprehensive review of the medical literature relevant to the topic. Members engaged in meetings and discussion through December 2021 to synthesize the information and arrive at a consensus. RESULTS: Existing infrastructure for robust AI development for AMD includes several large, labeled data sets of color fundus photography and OCT images; however, image data often do not contain the metadata necessary for the development of reliable, valid, and generalizable models. Data sharing for AMD model development is made difficult by restrictions on data privacy and security, although potential solutions are under investigation. Computing resources may be adequate for current applications, but knowledge of machine learning development may be scarce in many clinical ophthalmology settings. Despite these challenges, researchers have produced promising AI models for AMD for screening, diagnosis, prediction, and monitoring. Future goals include defining benchmarks to facilitate regulatory authorization and subsequent clinical setting generalization. CONCLUSIONS: Delivering an FDA-authorized, AI-based device for clinical care in AMD involves numerous considerations, including the identification of an appropriate clinical application; acquisition and development of a large, high-quality data set; development of the AI architecture; training and validation of the model; and functional interactions between the model output and clinical end user. The research efforts undertaken to date represent starting points for the medical devices that eventually will benefit providers, health care systems, and patients.

Pivotal Evaluation of an Artificial Intelligence System for Autonomous Detection of Referrable and Vision-Threatening Diabetic Retinopathy.

Importance: Diabetic retinopathy (DR) is a leading cause of blindness in adults worldwide. Early detection and intervention can prevent blindness; however, many patients do not receive their recommended annual diabetic eye examinations, primarily owing to limited access. Objective: To evaluate the safety and accuracy of an artificial intelligence (AI) system (the EyeArt Automated DR Detection System, version 2.1.0) in detecting both more-than-mild diabetic retinopathy (mtmDR) and vision-threatening diabetic retinopathy (vtDR). Design, Setting, and Participants: A prospective multicenter cross-sectional diagnostic study was preregistered (NCT03112005) and conducted from April 17, 2017, to May 30, 2018. A total of 942 individuals aged 18 years or older who had diabetes gave consent to participate at 15 primary care and eye care facilities. Data analysis was performed from February 14 to July 10, 2019. Interventions: Retinal imaging for the autonomous AI system and Early Treatment Diabetic Retinopathy Study (ETDRS) reference standard determination. Main Outcomes and Measures: Primary outcome measures included the sensitivity and specificity of the AI system in identifying participants' eyes with mtmDR and/or vtDR by 2-field undilated fundus photography vs a rigorous clinical reference standard comprising reading center grading of 4 wide-field dilated images using the ETDRS severity scale. Secondary outcome measures included the evaluation of imageability, dilated-if-needed analysis, enrichment correction analysis, worst-case imputation, and safety outcomes. Results: Of 942 consenting individuals, 893 patients (1786 eyes) met the inclusion criteria and completed the study protocol. The population included 449 men (50.3%). Mean (SD) participant age was 53.9 (15.2) years (median, 56; range, 18-88 years), 655 were White (73.3%), and 206 had type 1 diabetes (23.1%). Sensitivity and specificity of the AI system were high in detecting mtmDR (sensitivity: 95.5%; 95% CI, 92.4%-98.5% and specificity: 85.0%; 95% CI, 82.6%-87.4%) and vtDR (sensitivity: 95.1%; 95% CI, 90.1%-100% and specificity: 89.0%; 95% CI, 87.0%-91.1%) without dilation. Imageability was high without dilation, with the AI system able to grade 87.4% (95% CI, 85.2%-89.6%) of the eyes with reading center grades. When eyes with ungradable results were dilated per the protocol, the imageability improved to 97.4% (95% CI, 96.4%-98.5%), with the sensitivity and specificity being similar. After correcting for enrichment, the mtmDR specificity increased to 87.8% (95% CI, 86.3%-89.5%) and the sensitivity remained similar; for vtDR, both sensitivity (97.0%; 95% CI, 91.2%-100%) and specificity (90.1%; 95% CI, 89.4%-91.5%) improved. Conclusions and Relevance: This prospective multicenter cross-sectional diagnostic study noted safety and accuracy with use of the EyeArt Automated DR Detection System in detecting both mtmDR and, for the first time, vtDR, without physician assistance. These findings suggest that improved access to accurate, reliable diabetic eye examinations may increase adherence to recommended annual screenings and allow for accelerated referral of patients identified as having vtDR.

Imaging and artificial intelligence for progression of age-related macular degeneration.

Age-related macular degeneration (AMD) is a leading cause of severe vision loss. With our aging population, it may affect 288 million people globally by the year 2040. AMD progresses from an early and intermediate dry form to an advanced one, which manifests as choroidal neovascularization and geographic atrophy. Conversion to AMD-related exudation is known as progression to neovascular AMD, and presence of geographic atrophy is known as progression to advanced dry AMD. AMD progression predictions could enable timely monitoring, earlier detection and treatment, improving vision outcomes. Machine learning approaches, a subset of artificial intelligence applications, applied on imaging data are showing promising results in predicting progression. Extracted biomarkers, specifically from optical coherence tomography scans, are informative in predicting progression events. The purpose of this mini review is to provide an overview about current machine learning applications in artificial intelligence for predicting AMD progression, and describe the various methods, data-input types, and imaging modalities used to identify high-risk patients. With advances in computational capabilities, artificial intelligence applications are likely to transform patient care and management in AMD. External validation studies that improve generalizability to populations and devices, as well as evaluating systems in real-world clinical settings are needed to improve the clinical translations of artificial intelligence AMD applications.

Probabilistic Forecasting of Anti-VEGF Treatment Frequency in Neovascular Age-Related Macular Degeneration.

Purpose: To probabilistically forecast needed anti-vascular endothelial growth factor (anti-VEGF) treatment frequency using volumetric spectral domain-optical coherence tomography (SD-OCT) biomarkers in neovascular age-related macular degeneration from real-world settings. Methods: SD-OCT volume scans were segmented with a custom deep-learning-based analysis pipeline. Retinal thickness and reflectivity values were extracted for the central and the four inner Early Treatment Diabetic Retinopathy Study (ETDRS) subfields for six retinal layers (inner retina, outer nuclear layer, inner segments [IS], outer segments [OS], retinal pigment epithelium-drusen complex [RPEDC] and the choroid). Machine-learning models were probed to predict the anti-VEGF treatment frequency within the next 12 months. Probabilistic forecasting was performed using natural gradient boosting (NGBoost), which outputs a full probability distribution. The mean absolute error (MAE) between the predicted versus actual anti-VEGF treatment frequency was the primary outcome measure. Results: In a total of 138 visits of 99 eyes with neovascular AMD (96 patients) from two clinical centers, the prediction of future anti-VEGF treatment frequency was observed with an accuracy (MAE [95% confidence interval]) of 2.60 injections/year [2.25-2.96] (R2 = 0.390) using random forest regression and 2.66 injections/year [2.31-3.01] (R2 = 0.094) using NGBoost, respectively. Prediction intervals were well calibrated and reflected the true uncertainty of NGBoost-based predictions. Standard deviation of RPEDC-thickness in the central ETDRS-subfield constituted an important predictor across models. Conclusions: The proposed, fully automated pipeline enables probabilistic forecasting of future anti-VEGF treatment frequency in real-world settings. Translational Relevance: Prediction of a probability distribution allows the physician to inspect the underlying uncertainty. Predictive uncertainty estimates are essential to highlight cases where human-inspection and/or reversion to a fallback alternative is warranted.