Visualizing features with wide-field volumetric OCT angiography.

Optical coherence tomography (OCT) and its extension OCT angiography (OCTA) have become essential clinical imaging modalities due to their ability to provide depth-resolved angiographic and tissue structural information non-invasively and at high resolution. Within a field of view, the anatomic detail available is sufficient to identify several structural and vascular pathologies that are clinically relevant for multiple prevalent blinding diseases, including age-related macular degeneration (AMD), diabetic retinopathy (DR), and vein occlusions. The main limitation in contemporary OCT devices is that this field of view is limited due to a fundamental trade-off between system resolution/sensitivity, sampling density, and imaging window dimensions. Here, we describe a swept-source OCT device that can capture up to a 12 × 23-mm field of view in a single shot and show that it can identify conventional pathologic features such as non-perfusion areas outside of conventional fields of view. We also show that our approach maintains sensitivity sufficient to visualize novel features, including choriocapillaris morphology beneath the macula and macrophage-like cells at the inner limiting membrane, both of which may have implications for disease.

Bias Reduction Practices in Underrepresented Groups in Ophthalmology Resident Recruitment.

Importance: Best recruitment practices for increasing diversity are well established, but the adoption and impact of these practices in ophthalmology residency recruitment are unknown. Objective: To describe the adoption of bias reduction practices in groups underrepresented in ophthalmology (URiO) residency recruitment and determine which practices are effective for increasing URiO residents. Design, Setting, and Participants: This cross-sectional survey study used an 18-item questionnaire included in the online survey of the Association of University Professors in Ophthalmology (AUPO) Residency Program Directors. Data collection occurred from July 2022 to December 2022. The data were initially analyzed on January 16, 2023. Participants included residency program directors (PDs) in the AUPO PD listserv database. Main Outcomes and Measures: Descriptive analysis of resident selection committee approaches, evaluation of applicant traits, and use of bias reduction tools. Primary outcome was diversity assessed by presence of at least 1 resident in the last 5 classes who identified as URiO, including those underrepresented in medicine (URiM), lesbian, gay, bisexual, transgender, queer, intersex, and asexual plus, or another disadvantaged background (eg, low socioeconomic status). Multivariate analyses of recruitment practices were conducted to determine which practices were associated with increased URiO and URiM. Results: Among 106 PDs, 65 completed the survey (61.3%). Thirty-nine PDs used an interview rubric (60.0%), 28 used interview standardization (43.0%), 56 provided at least 1 bias reduction tool to their selection committee (86.2%), and 44 used postinterview metrics to assess diversity, equity, and inclusion efforts (67.7%). Application filters, interview standardization, and postinterview metrics were not associated with increased URiO. Multivariate logistic regression analysis showed larger residency class (odds ratio [OR], 1.34; 95% CI, 1.09-1.65; P = .01) and use of multiple selection committee bias reduction tools (OR, 1.47; 95% CI, 1.13-1.92; P = .01) were positively associated with increased URiO, whereas use of interview rubrics (OR, 0.72; 95% CI, 0.59-0.87; P = .001) and placing higher importance of applicant interest in a program (OR, 0.83; 95% CI, 0.75-0.92; P = .02) were negatively associated. URiM analyses showed similar associations. Conclusions and Relevance: Ophthalmology residency interviews are variably standardized. In this study, providing multiple bias reduction tools to selection committees was associated with increased URiO and URiM residents. Prioritizing applicant interest in a program may reduce resident diversity. Interview rubrics, while intended to reduce bias, may inadvertently increase inequity.

Interpretable Diabetic Retinopathy Diagnosis Based on Biomarker Activation Map.

OBJECTIVE: Deep learning classifiers provide the most accurate means of automatically diagnosing diabetic retinopathy (DR) based on optical coherence tomography (OCT) and its angiography (OCTA). The power of these models is attributable in part to the inclusion of hidden layers that provide the complexity required to achieve a desired task. However, hidden layers also render algorithm outputs difficult to interpret. Here we introduce a novel biomarker activation map (BAM) framework based on generative adversarial learning that allows clinicians to verify and understand classifiers' decision-making. METHODS: A data set including 456 macular scans were graded as non-referable or referable DR based on current clinical standards. A DR classifier that was used to evaluate our BAM was first trained based on this data set. The BAM generation framework was designed by combing two U-shaped generators to provide meaningful interpretability to this classifier. The main generator was trained to take referable scans as input and produce an output that would be classified by the classifier as non-referable. The BAM is then constructed as the difference image between the output and input of the main generator. To ensure that the BAM only highlights classifier-utilized biomarkers an assistant generator was trained to do the opposite, producing scans that would be classified as referable by the classifier from non-referable scans. RESULTS: The generated BAMs highlighted known pathologic features including nonperfusion area and retinal fluid. CONCLUSION/SIGNIFICANCE: A fully interpretable classifier based on these highlights could help clinicians better utilize and verify automated DR diagnosis.

Perfused and Nonperfused Microaneurysms Identified and Characterized by Structural and Angiographic OCT.

PURPOSE: Microaneurysms (MAs) have distinct, oval-shaped, hyperreflective walls on structural OCT, and inconsistent flow signal in the lumen with OCT angiography (OCTA). Their relationship to regional macular edema in diabetic retinopathy (DR) has not been quantitatively explored. DESIGN: Retrospective, cross-sectional study. PARTICIPANTS: A total of 99 participants, including 23 with mild, nonproliferative DR (NPDR), 25 with moderate NPDR, 34 with severe NPDR, and 17 with proliferative DR. METHODS: We obtained 3 × 3-mm scans with a commercial device (Solix, Visionix/Optovue) in 99 patients with DR. Trained graders manually identified MAs and their location relative to the anatomic layers from cross-sectional OCT. Microaneurysms were first classified as perfused if flow signal was present in the OCTA channel. Then, perfused MAs were further classified into fully and partially perfused MAs based on the flow characteristics in en face OCTA. The presence of retinal fluid based on OCT near MAs was compared between perfused and nonperfused types. We also compared OCT-based MA detection to fundus photography (FP)- and fluorescein angiography (FA)-based detection. MAIN OUTCOME MEASURES: OCT-identified MAs can be classified according to colocalized OCTA flow signal into fully perfused, partially perfused, and nonperfused types. Fully perfused MAs may be more likely to be associated with diabetic macular edema (DME) than those without flow. RESULTS: We identified 308 MAs (166 fully perfused, 88 partially perfused, 54 nonperfused) in 42 eyes using OCT and OCTA. Nearly half of the MAs identified in this study straddle the inner nuclear layer and outer plexiform layer. Compared with partially perfused and nonperfused MAs, fully perfused MAs were more likely to be associated with local retinal fluid. The associated fluid volumes were larger with fully perfused MAs compared with other types. OCT/OCTA detected all MAs found on FP. Although not all MAs seen with FA were identified with OCT, some MAs seen with OCT were not visible with FA or FP. CONCLUSIONS: OCT-identified MAs with colocalized flow on OCTA are more likely to be associated with DME than those without flow. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

The Impact of Documentation Workflow on the Accuracy of the Coded Diagnoses in the Electronic Health Record.

Objective: To determine the impact of documentation workflow on the accuracy of coded diagnoses in electronic health records (EHRs). Design: Cross-sectional study. Participants: All patients who completed visits at the Casey Eye Institute Retina Division faculty clinic between April 7, 2022 and April 13, 2022. Main Outcome Measures: Agreement between coded diagnoses and clinical notes. Methods: We assessed the rate of agreement between the diagnoses in the clinical notes and the coded diagnosis in the EHR using manual review and examined the impact of the documentation workflow on the rate of agreement in an academic retina practice. Results: In 202 visits by 8 physicians, 78% (range, 22%-100%) had an agreement between the coded diagnoses and the clinical notes. When physicians integrated the diagnosis code entry and note composition, the rate of agreement was 87.9% (range, 62%-100%). For those who entered the diagnosis codes separately from writing notes, the agreement was 44.4% (22%-50%, P < 0.0001). Conclusion: The visit-specific agreement between the coded diagnosis and the progress note can vary widely by workflow. The workflow and EHR design may be an important part of understanding and improving the quality of EHR data. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Early Sign of Retinal Neovascularization Evolution in Diabetic Retinopathy: A Longitudinal OCT Angiography Study.

Purpose: To assess whether the combination of en face OCT and OCT angiography (OCTA) can capture observable, but subtle, structural changes that precede clinically evident retinal neovascularization (RNV) in eyes with diabetic retinopathy (DR). Design: Retrospective, longitudinal study. Participants: Patients with DR that had at least 2 visits. Methods: We obtained wide-field OCTA scans of 1 eye from each participant and generated en face OCT, en face OCTA, and cross-sectional OCTA. We identified eyes with RNV sprouts, defined as epiretinal hyperreflective materials on en face OCT with flow signals breaching the internal limiting membrane on the cross-sectional OCTA without recognizable RNV on en face OCTA and RNV fronds, defined as recognizable abnormal vascular structures on the en face OCTA. We examined the corresponding location from follow-up or previous visits for the presence or progression of the RNV. Main Outcome Measures: The characteristics and longitudinal observation of early signs of RNV. Results: From 71 eyes, we identified RNV in 20 eyes with the combination of OCT and OCTA, of which 13 (65%) were photographically graded as proliferative DR, 6 (30%) severe nonproliferative DR, and 1 (5%) moderate nonproliferative diabetic retinopathy. From these eyes, we identified 38 RNV sprouts and 26 RNV fronds at the baseline. Thirty-four RNVs (53%) originated from veins, 24 (38%) were from intraretinal microabnormalities, and 6 (9%) were from a nondilated capillary bed. At the final visit, 53 RNV sprouts and 30 RNV fronds were detected. Ten eyes (50%) showed progression, defined as having a new RNV lesion or the development of an RNV frond from an RNV sprout. Four (11%) RNV sprouts developed into RNV fronds with a mean interval of 7.0 months. Nineteen new RNV sprouts developed during the follow-up, whereas no new RNV frond was observed outside an identified RNV sprout. The eyes with progression were of younger age (P = 0.014) and tended to be treatment naive (P = 0.07) compared with eyes without progression. Conclusions: Longitudinal observation demonstrated that a combination of en face OCT and cross-sectional OCTA can identify an earlier form of RNV before it can be recognized on en face OCTA. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Perfused and Nonperfused Microaneurysms Identified and Characterized by Structural and Angiographic OCT.

Purpose: Microaneurysms (MAs) have distinct, oval-shaped, hyperreflective walls on structural OCT, and inconsistent flow signal in the lumen with OCT angiography (OCTA). Their relationship to regional macular edema in diabetic retinopathy (DR) has not been quantitatively explored. Design: Retrospective, cross-sectional study. Participants: A total of 99 participants, including 23 with mild, nonproliferative DR (NPDR), 25 with moderate NPDR, 34 with severe NPDR, and 17 with proliferative DR. Methods: We obtained 3 × 3-mm scans with a commercial device (Solix, Visionix/Optovue) in 99 patients with DR. Trained graders manually identified MAs and their location relative to the anatomic layers from cross-sectional OCT. Microaneurysms were first classified as perfused if flow signal was present in the OCTA channel. Then, perfused MAs were further classified into fully and partially perfused MAs based on the flow characteristics in en face OCTA. The presence of retinal fluid based on OCT near MAs was compared between perfused and nonperfused types. We also compared OCT-based MA detection to fundus photography (FP)- and fluorescein angiography (FA)-based detection. Main Outcome Measures: OCT-identified MAs can be classified according to colocalized OCTA flow signal into fully perfused, partially perfused, and nonperfused types. Fully perfused MAs may be more likely to be associated with diabetic macular edema (DME) than those without flow. Results: We identified 308 MAs (166 fully perfused, 88 partially perfused, 54 nonperfused) in 42 eyes using OCT and OCTA. Nearly half of the MAs identified in this study straddle the inner nuclear layer and outer plexiform layer. Compared with partially perfused and nonperfused MAs, fully perfused MAs were more likely to be associated with local retinal fluid. The associated fluid volumes were larger with fully perfused MAs compared with other types. OCT/OCTA detected all MAs found on FP. Although not all MAs seen with FA were identified with OCT, some MAs seen with OCT were not visible with FA or FP. Conclusions: OCT-identified MAs with colocalized flow on OCTA are more likely to be associated with DME than those without flow. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article. Ophthalmology Retina 2023;■:1-8 © 2023 by the American Academy of Ophthalmology. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Automated Macular Fluid Volume As a Treatment Indicator for Diabetic Macular Edema.

Introduction: To assess the diagnostic accuracy of automatically quantified macular fluid volume (MFV) for treatment-required diabetic macular edema (DME). Methods: This retrospective cross-sectional study included eyes with DME. The commercial software on optical coherence tomography (OCT) produced the central subfield thickness (CST), and a custom deep-learning algorithm automatically segmented the fluid cysts and quantified the MFV from the volumetric scans of an OCT angiography system. Retina specialists treated patients per standard of care based on clinical and OCT findings without access to the MFV. The main outcome measures were the area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity of the CST, MFV, and visual acuity (VA) for treatment indication. Results: Of 139 eyes, 39 (28%) were treated for DME during the study period and 101 (72%) were previously treated. The algorithm detected fluid in all eyes; however, only 54 eyes (39%) met the DRCR.net criteria for center-involved ME. The AUROC of MFV predicting a treatment decision of 0.81 was greater than that of CST (0.67) (P = .0048). Untreated eyes that met the optimal threshold for treatment-required DME based on MFV (>0.031 mm3) had better VA than treated eyes (P = .0053). A multivariate logistic regression model showed that MFV (P = .0008) and VA (P = .0061) were significantly associated with a treatment decision, but CST was not. Conclusions: MFV had a higher correlation with the need for treatment for DME than CST and may be especially useful for ongoing management of DME.

Signal attenuation-compensated projection-resolved OCT angiography.

Projection artifacts are a significant limitation of optical coherence tomographic angiography (OCTA). Existing techniques to suppress these artifacts are sensitive to image quality, becoming less reliable on low-quality images. In this study, we propose a novel signal attenuation-compensated projection-resolved OCTA (sacPR-OCTA) algorithm. In addition to removing projection artifacts, our method compensates for shadows beneath large vessels. The proposed sacPR-OCTA algorithm improves vascular continuity, reduces the similarity of vascular patterns in different plexuses, and removes more residual artifacts compared to existing methods. In addition, the sacPR-OCTA algorithm better preserves flow signal in choroidal neovascular lesions and shadow-affected areas. Because sacPR-OCTA processes the data along normalized A-lines, it provides a general solution for removing projection artifacts agnostic to the platform.

Utility of En Face OCT for the Detection of Clinically Unsuspected Retinal Neovascularization in Patients with Diabetic Retinopathy.

PURPOSE: To assess the value of en face OCT for detecting clinically unsuspected retinal neovascularization (RNV) in patients with nonproliferative diabetic retinopathy (NPDR). DESIGN: A retrospective, cross-sectional study. PARTICIPANTS: Treatment-naïve patients clinically graded as NPDR in an ongoing prospective observational OCT angiography (OCTA) study at a tertiary care center. METHODS: Each patient underwent imaging of 1 eye with a spectral-domain OCTA, generating a 17 × 17-mm widefield image by montaging four 9 × 9-mm scans. Two independent graders examined a combination of en face OCT, en face OCTA with a custom vitreoretinal interface slab, and cross-sectional OCTA to determine the presence of RNV. We measured the area of RNV flow within RNV lesions on en face OCTA. MAIN OUTCOME MEASURES: Detection rate of clinically occult RNV with OCT and OCTA. RESULTS: Of 63 enrolled eyes, 27 (43%) were clinically graded as severe NPDR, 16 (25%) as moderate NPDR, and 20 (32%) as mild NPDR. Using the combination of en face OCT, en face OCTA, and cross-sectional OCTA, the graders detected 42 RNV lesions in 12 (19%) eyes, of which 8 (67%) were graded as severe NPDR, 2 (17%) as moderate NPDR, and 2 (17%) as mild NPDR. The sensitivity of en face OCT alone for detecting eyes with RNV was similar to that of en face OCTA alone (100% vs. 92%; P = 0.32), whereas the specificity of en face OCT alone was significantly lower than that of en face OCTA alone (32% vs. 73%; P < 0.001). For detecting individual RNV lesions, the en face OCT was 100% sensitive, compared with 67% sensitivity for the en face OCTA (P < 0.001). The area of RNV lesions that manual grading with en face OCTA alone missed was significantly smaller than that of manually detectable RNV (Mean [standard deviation] RNV flow area, 0.015 [0.020] mm2 vs. 0.16 [0.36] mm2; P < 0.001). CONCLUSION: The combination of en face OCT and OCTA can detect clinically occult RNV with high sensitivity. For screening these small lesions, en face OCT may be a useful imaging modality. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found after the references.

Deep-Learning-Aided Diagnosis of Diabetic Retinopathy, Age-Related Macular Degeneration, and Glaucoma Based on Structural and Angiographic OCT.

Purpose: Timely diagnosis of eye diseases is paramount to obtaining the best treatment outcomes. OCT and OCT angiography (OCTA) have several advantages that lend themselves to early detection of ocular pathology; furthermore, the techniques produce large, feature-rich data volumes. However, the full clinical potential of both OCT and OCTA is stymied when complex data acquired using the techniques must be manually processed. Here, we propose an automated diagnostic framework based on structural OCT and OCTA data volumes that could substantially support the clinical application of these technologies. Design: Cross sectional study. Participants: Five hundred twenty-six OCT and OCTA volumes were scanned from the eyes of 91 healthy participants, 161 patients with diabetic retinopathy (DR), 95 patients with age-related macular degeneration (AMD), and 108 patients with glaucoma. Methods: The diagnosis framework was constructed based on semisequential 3-dimensional (3D) convolutional neural networks. The trained framework classifies combined structural OCT and OCTA scans as normal, DR, AMD, or glaucoma. Fivefold cross-validation was performed, with 60% of the data reserved for training, 20% for validation, and 20% for testing. The training, validation, and test data sets were independent, with no shared patients. For scans diagnosed as DR, AMD, or glaucoma, 3D class activation maps were generated to highlight subregions that were considered important by the framework for automated diagnosis. Main Outcome Measures: The area under the curve (AUC) of the receiver operating characteristic curve and quadratic-weighted kappa were used to quantify the diagnostic performance of the framework. Results: For the diagnosis of DR, the framework achieved an AUC of 0.95 ± 0.01. For the diagnosis of AMD, the framework achieved an AUC of 0.98 ± 0.01. For the diagnosis of glaucoma, the framework achieved an AUC of 0.91 ± 0.02. Conclusions: Deep learning frameworks can provide reliable, sensitive, interpretable, and fully automated diagnosis of eye diseases. Financial Disclosures: Proprietary or commercial disclosure may be found after the references.

Deep learning-based signal-independent assessment of macular avascular area on 6×6 mm optical coherence tomography angiogram in diabetic retinopathy: a comparison to instrument-embedded software.

SYNOPSIS: A deep-learning-based macular extrafoveal avascular area (EAA) on a 6×6 mm optical coherence tomography (OCT) angiogram is less dependent on the signal strength and shadow artefacts, providing better diagnostic accuracy for diabetic retinopathy (DR) severity than the commercial software measured extrafoveal vessel density (EVD). AIMS: To compare a deep-learning-based EAA to commercial output EVD in the diagnostic accuracy of determining DR severity levels from 6×6 mm OCT angiography (OCTA) scans. METHODS: The 6×6 mm macular OCTA scans were acquired on one eye of each participant with a spectral-domain OCTA system. After excluding the central 1 mm diameter circle, the EAA on superficial vascular complex was measured with a deep-learning-based algorithm, and the EVD was obtained with commercial software. RESULTS: The study included 34 healthy controls and 118 diabetic patients. EAA and EVD were highly correlated with DR severity (ρ=0.812 and -0.577, respectively, both p<0.001) and visual acuity (r=-0.357 and 0.420, respectively, both p<0.001). EAA had a significantly (p<0.001) higher correlation with DR severity than EVD. With the specificity at 95%, the sensitivities of EAA for differentiating diabetes mellitus (DM), DR and severe DR from control were 80.5%, 92.0% and 100.0%, respectively, significantly higher than those of EVD 11.9% (p=0.001), 13.6% (p<0.001) and 15.8% (p<0.001), respectively. EVD was significantly correlated with signal strength index (SSI) (r=0.607, p<0.001) and shadow area (r=-0.530, p<0.001), but EAA was not (r=-0.044, p=0.805 and r=-0.046, p=0.796, respectively). Adjustment of EVD with SSI and shadow area lowered sensitivities for detection of DM, DR and severe DR. CONCLUSION: Macular EAA on 6×6 mm OCTA measured with a deep learning-based algorithm is less dependent on the signal strength and shadow artefacts, and provides better diagnostic accuracy for DR severity than EVD measured with the instrument-embedded software.

A Deep Learning Network for Classifying Arteries and Veins in Montaged Widefield OCT Angiograms.

Purpose: To propose a deep-learning-based method to differentiate arteries from veins in montaged widefield OCT angiography (OCTA). Design: Cross-sectional study. Participants: A total of 232 participants, including 109 participants with diabetic retinopathy (DR), 64 participants with branch retinal vein occlusion (BRVO), 27 participants with diabetes but without DR, and 32 healthy participants. Methods: We propose a convolutional neural network (CAVnet) to classify retinal blood vessels on montaged widefield OCTA en face images as arteries and veins. A total of 240 retinal angiograms from 88 eyes were used to train CAVnet, and 302 retinal angiograms from 144 eyes were used for testing. This method takes the OCTA images as input and outputs the segmentation results with arteries and veins down to the level of precapillary arterioles and postcapillary venules. The network also identifies their intersections. We evaluated the agreement (in pixels) between segmentation results and the manually graded ground truth using sensitivity, specificity, F1-score, and Intersection over Union (IoU). Measurements of arterial and venous caliber or tortuosity are made on our algorithm's output of healthy and diseased eyes. Main Outcome Measures: Classification of arteries and veins, arterial and venous caliber, and arterial and venous tortuosity. Results: For classification and identification of arteries, the algorithm achieved average sensitivity of 95.3%, specificity of 99.6%, F1 score of 94.2%, and IoU of 89.3%. For veins, the algorithm achieved average sensitivity of 94.4%, specificity of 99.7%, F1 score of 94.1%, and IoU of 89.2%. We also achieved an average sensitivity of 76.3% in identifying intersection points. The results show CAVnet has high accuracy on differentiating arteries and veins in DR and BRVO cases. These classification results are robust across 2 instruments and multiple scan volume sizes. Outputs of CAVnet were used to measure arterial and venous caliber or tortuosity, and pixel-wise caliber and tortuosity maps were generated. Differences between healthy and diseased eyes were demonstrated, indicating potential clinical utility. Conclusions: The CAVnet can classify arteries and veins and their branches with high accuracy and is potentially useful in the analysis of vessel type-specific features on diseases such as branch retinal artery occlusion and BRVO.

Volume-based, layer-independent, disease-agnostic detection of abnormal retinal reflectivity, nonperfusion, and neovascularization using structural and angiographic OCT.

Optical coherence tomography (OCT) is widely used in ophthalmic practice because it can visualize retinal structure and vasculature in vivo and 3-dimensionally (3D). Even though OCT procedures yield data volumes, clinicians typically interpret the 3D images using two-dimensional (2D) data subsets, such as cross-sectional scans or en face projections. Since a single OCT volume can contain hundreds of cross-sections (each of which must be processed with retinal layer segmentation to produce en face images), a thorough manual analysis of the complete OCT volume can be prohibitively time-consuming. Furthermore, 2D reductions of the full OCT volume may obscure relationships between disease progression and the (volumetric) location of pathology within the retina and can be prone to mis-segmentation artifacts. In this work, we propose a novel framework that can detect several retinal pathologies in three dimensions using structural and angiographic OCT. Our framework operates by detecting deviations in reflectance, angiography, and simulated perfusion from a percent depth normalized standard retina created by merging and averaging scans from healthy subjects. We show that these deviations from the standard retina can highlight multiple key features, while the depth normalization obviates the need to segment several retinal layers. We also construct a composite pathology index that measures average deviation from the standard retina in several categories (hypo- and hyper-reflectance, nonperfusion, presence of choroidal neovascularization, and thickness change) and show that this index correlates with DR severity. Requiring minimal retinal layer segmentation and being fully automated, this 3D framework has a strong potential to be integrated into commercial OCT systems and to benefit ophthalmology research and clinical care.

THREE-DIMENSIONAL QUANTIFICATION OF INTRARETINAL CYSTOID SPACES ASSOCIATED WITH FULL-THICKNESS MACULAR HOLE.

PURPOSE: To evaluate intraretinal cystoid spaces in patients with idiopathic macular hole (MH). METHODS: Retrospective cohort study included consecutive patients with full-thickness MH who underwent successful MH surgery and 12 months of follow-up. Custom software was applied to preoperative optical coherence tomography scans to generate fluid volume. Inner fluid volume was defined as cystoid spaces in the inner nuclear layer, and outer fluid volume was defined as cystoid spaces in Henle fiber layer of the outer nuclear layer. RESULTS: Thirty-nine eyes from 39 participants were included. Postoperative 12-month visual acuity correlated with both inner fluid volume and minimum MH size (both P < 0.05) but not outer fluid volume. Inner fluid volume positively correlated with minimum MH size ( P = 0.0003). After accounting for minimum MH size with multivariable analysis, inner fluid volume effect on VA remained significant ( P = 0.025). After dividing inner fluid volume into tertiles, mean baseline visual acuity was 20/50 in eyes with small inner fluid volume, and was 20/125 in eyes with large inner fluid volume ( P = 0.0039). Mean postoperative 12-month visual acuity was 20/20 in eyes with small inner fluid volume compared with 20/32 in eyes with large inner fluid volume ( P = 0.019). CONCLUSION: Increased inner fluid volume was associated with worse postoperative VA.

A Diabetic Retinopathy Classification Framework Based on Deep-Learning Analysis of OCT Angiography.

Purpose: Reliable classification of referable and vision threatening diabetic retinopathy (DR) is essential for patients with diabetes to prevent blindness. Optical coherence tomography (OCT) and its angiography (OCTA) have several advantages over fundus photographs. We evaluated a deep-learning-aided DR classification framework using volumetric OCT and OCTA. Methods: Four hundred fifty-six OCT and OCTA volumes were scanned from eyes of 50 healthy participants and 305 patients with diabetes. Retina specialists labeled the eyes as non-referable (nrDR), referable (rDR), or vision threatening DR (vtDR). Each eye underwent a 3 × 3-mm scan using a commercial 70 kHz spectral-domain OCT system. We developed a DR classification framework and trained it using volumetric OCT and OCTA to classify eyes into rDR and vtDR. For the scans identified as rDR or vtDR, 3D class activation maps were generated to highlight the subregions which were considered important by the framework for DR classification. Results: For rDR classification, the framework achieved a 0.96 ± 0.01 area under the receiver operating characteristic curve (AUC) and 0.83 ± 0.04 quadratic-weighted kappa. For vtDR classification, the framework achieved a 0.92 ± 0.02 AUC and 0.73 ± 0.04 quadratic-weighted kappa. In addition, the multiple DR classification (non-rDR, rDR but non-vtDR, or vtDR) achieved a 0.83 ± 0.03 quadratic-weighted kappa. Conclusions: A deep learning framework only based on OCT and OCTA can provide specialist-level DR classification using only a single imaging modality. Translational Relevance: The proposed framework can be used to develop clinically valuable automated DR diagnosis system because of the specialist-level performance showed in this study.

Plexus-specific retinal capillary avascular area in exudative age-related macular degeneration with projection-resolved OCT angiography.

OBJECTIVE: To detect the plexus-specific retinal capillary avascular area in exudative age-related macular degeneration (EAMD) with projection-resolved optical coherence tomography angiography (PR-OCTA). METHODS AND ANALYSIS: In this prospective cross-sectional single centre study, eyes with treatment-naïve EAMD underwent macular 3×3 mm OCTA with AngioVue system. OCTA scans were analysed and processed including three-dimensional projection artefact removal, retinal layer semi-automated segmentation and en face angiogram generation. Automated quantification of extrafoveal (excluding the central 1 mm circle) avascular area (EAA) were calculated on projection-resolved superficial vascular complex (SVC), intermediate capillary plexus (ICP) and deep capillary plexus (DCP), respectively. RESULTS: Nineteen eyes with EAMD and 19 age-matched healthy control eyes were included. There was no significant difference between the EAMD and control eyes in terms of age, sex, axial length and mean ocular perfusion pressure (all p>0.05). Compared with control eyes, EAMD eyes had significantly larger EAA in SVC (median 0.125 vs 0.059 mm2, p=0.006), ICP (0.016 vs 0.000 mm2, p=0.004) and DCP (0.033 vs 0.000 mm2, p＜0.001). CONCLUSION: PR-OCTA showed that EAMD is associated with focal avascular area in all the three retinal vascular plexuses.

Association Between Fluid Volume in Inner Nuclear Layer and Visual Acuity in Diabetic Macular Edema.

PURPOSE: In diabetic macular edema (DME), the correlation between visual acuity (VA) and central subfield thickness (CST) is weak. We hypothesize that fluid volume (FV) in the inner nuclear layer (INL) may correlate more strongly with VA. DESIGN: Retrospective, cross-sectional study. METHODS: One eye each of diabetic patients with DME was included. We measured intraretinal fluid volume that was detected by automated fluid detection algorithm on 3- × 3-mm optical coherence tomography angiogram volume scans. The detected fluid was subdivided into inner FV, bounded by the INL, and outer FV, the fluid between the outer border of INL to the ellipsoid zone. RESULTS: We enrolled 125 patients with DME (60 women; mean age, 61 years). The mean detected inner FV was 0.013 mm3 in 109 eyes (87%). The mean detected outer FV was 0.042 mm3 in 124 eyes (99%). Univariate analysis demonstrated that the VA significantly correlated with the inner FV (P < .0001), whole macular FV (P = .010), and CST (P = .036). Multivariate analysis demonstrated that the inner FV was the only significant factor (ß = -0.41, P = .004). These correlations were consistent when the treatment-naïve group (n = 33) and the eyes without previous laser treatments (n = 93) were analyzed separately. The area under the receiver operating characteristic curve of inner FV for VA of 20/32 or worse was significantly higher than that for CST (0.66 vs 0.54, P = .018). CONCLUSIONS: The inner FV has a stronger association with VA than other OCT biomarkers in DME and may be more clinically useful.

Geographic Atrophy Progression Is Associated With Choriocapillaris Flow Deficits Measured With Optical Coherence Tomographic Angiography.

Purpose: The purpose of this study was to assess the associations between baseline choriocapillaris (CC) flow deficits and geographic atrophy (GA) progression. Methods: In this prospective cohort study, patients with GA underwent 3 × 3-mm macular spectral-domain optical coherence tomographic angiography (OCTA) at baseline and follow-up visits. Annual GA enlargement rate was defined as change of square root of GA area in 12 months. Shadow areas due to iris, media opacity, retinal vessels, and drusen were excluded. CC vessel density (CC-VD) in non-GA areas was measured using a validated machine-learning-based algorithm. Low perfusion area (LPA) was defined as capillary density below the 0.1 percentile threshold of the same location of 40 normal healthy control eye. Focal perfusion loss (FPL) was defined as percentage of CC loss within LPA compared with normal controls. Results: Ten patients with GA were enrolled and followed for 26 months on average. At baseline, the mean GA area was 0.84 ± 0.70 mm2. The mean CC-VD was 44.5 ± 15.2%, the mean LPA was 4.29 ± 2.6 mm2, and the mean FPL was 50.4 ± 28.2%. The annual GA enlargement rate was significantly associated with baseline CC-VD (r = -0.816, P = 0.004), LPA (r = 0.809, P = 0.005), and FPL (r = 0.800, P = 0.005), but not with age (r = 0.008, P = 0.98) and GA area (r = -0.362, P = 0.30). Conclusions: Baseline CC flow deficits were significantly associated with a faster GA enlargement over the course of 1 year, suggesting the choriocapillaris perfusion outside of a GA area may play a role in GA progression.

An Open-Source Deep Learning Network for Reconstruction of High-Resolution OCT Angiograms of Retinal Intermediate and Deep Capillary Plexuses.

Purpose: We propose a deep learning-based image reconstruction algorithm to produce high-resolution optical coherence tomographic angiograms (OCTA) of the intermediate capillary plexus (ICP) and deep capillary plexus (DCP). Methods: In this study, 6-mm × 6-mm macular scans with a 400 × 400 A-line sampling density and 3-mm × 3-mm scans with a 304 × 304 A-line sampling density were acquired on one or both eyes of 180 participants (including 230 eyes with diabetic retinopathy and 44 healthy controls) using a 70-kHz commercial OCT system (RTVue-XR; Optovue, Inc., Fremont, California, USA). Projection-resolved OCTA algorithm removed projection artifacts in voxel. ICP and DCP angiograms were generated by maximum projection of the OCTA signal within the respective plexus. We proposed a deep learning-based method, which receives inputs from registered 3-mm × 3-mm ICP and DCP angiograms with proper sampling density as the ground truth reference to reconstruct 6-mm × 6-mm high-resolution ICP and DCP en face OCTA. We applied the same network on 3-mm × 3-mm angiograms to enhance these images further. We evaluated the reconstructed 3-mm × 3-mm and 6-mm × 6-mm angiograms based on vascular connectivity, Weber contrast, false flow signal (flow signal erroneously generated from background), and the noise intensity in the foveal avascular zone. Results: Compared to the originals, the Deep Capillary Angiogram Reconstruction Network (DCARnet)-enhanced 6-mm × 6-mm angiograms had significantly reduced noise intensity (ICP, 7.38 ± 25.22, P < 0.001; DCP, 11.20 ± 22.52, P < 0.001), improved vascular connectivity (ICP, 0.95 ± 0.01, P < 0.001; DCP, 0.96 ± 0.01, P < 0.001), and enhanced Weber contrast (ICP, 4.25 ± 0.10, P < 0.001; DCP, 3.84 ± 0.84, P < 0.001), without generating false flow signal when noise intensity lower than 650. The DCARnet-enhanced 3-mm × 3-mm angiograms also reduced noise, improved connectivity, and enhanced Weber contrast in 3-mm × 3-mm ICP and DCP angiograms from 101 eyes. In addition, DCARnet preserved the appearance of the dilated vessels in the reconstructed angiograms in diabetic eyes. Conclusions: DCARnet can enhance 3-mm × 3-mm and 6-mm × 6-mm ICP and DCP angiogram image quality without introducing artifacts. Translational Relevance: The enhanced 6-mm × 6-mm angiograms may be easier for clinicians to interpret qualitatively.