Objective: To evaluate the accuracy and feasibility of the auto-detection of 15 retinal disorders with artificial intelligence (AI)-assisted optical coherence tomography (OCT) in community screening.Methods: A total of 954 eyes of 477 subjects from four local communities were enrolled in this study from September to December 2021. They received OCT scans covering an area of 12 mm × 9 mm at the posterior pole retina involving the macular and optic disc, as well as other ophthalmic examinations performed using their demographic information recorded. The OCT images were analyzed using integrated software with the previously established algorithm based on the deep-learning method and trained to detect 15 kinds of retinal disorders, namely, pigment epithelial detachment (PED), posterior vitreous detachment (PVD), epiretinal membranes (ERMs), sub-retinal fluid (SRF), choroidal neovascularization (CNV), drusen, retinoschisis, cystoid macular edema (CME), exudation, macular hole (MH), retinal detachment (RD), ellipsoid zone disruption, focal choroidal excavation (FCE), choroid atrophy, and retinal hemorrhage. Meanwhile, the diagnosis was also generated from three groups of individual ophthalmologists (group of retina specialists, senior ophthalmologists, and junior ophthalmologists) and compared with those by the AI. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were calculated, and kappa statistics were performed.Results: A total of 878 eyes were finally enrolled, with 76 excluded due to poor image quality. In the detection of 15 retinal disorders, the ROC curve comparison between AI and professors’ presented relatively large AUC (0.891–0.997), high sensitivity (87.65–100%), and high specificity (80.12–99.41%). Among the ROC curve comparisons with those by the retina specialists, AI was the closest one to the professors’ compared to senior and junior ophthalmologists (p < 0.05).Conclusion: AI-assisted OCT is highly accurate, sensitive, and specific in auto-detection of 15 kinds of retinal disorders, certifying its feasibility and effectiveness in community ophthalmic screening.

Objective: To detect fundus changes in patients with cerebral small vessel disease (CSVD) using optical coherence tomography angiography (OCTA) and to investigate the correlations between CSVD and fundus changes.Methods: From January 2019 to January 2020, patients diagnosed with CSVD by magnetic resonance imaging (MRI) were enrolled in our study and received fundus examinations using OCTA. CSVD was defined as white matter hyperintensities, enlarged perivascular spaces, lacunes, or microbleeds on MRI. OCTA parameters included foveal avascular zone areas, retinal nerve fiber layer thickness, and capillary densities of the superficial retinal capillary plexuses, deep retinal capillary plexuses, and the radial peripapillary capillary network of the disc. Univariate and multivariate logistic regression analyses were performed to explore the correlation between CSVD and fundus changes.Results: A total of 115 patients (40% male) were enrolled and analyzed, and the mean age was 65.11 ± 11.23 years. After multivariate logistic regression analysis, the radial peripapillary capillary network density was negatively correlated with severity of deep white matter lesions (OR: 0.909; 95% CI: 0.828–0.998; p = 0.046) and perivascular spaces (OR: 0.881; 95% CI: 0.779–0.995; p = 0.041). Parafoveal vessel densities of the superficial retinal capillary plexuses were independently correlated with lacunes (OR: 0.889; 95% CI: 0.817–0.967; p = 0.006).Conclusion: OCTA parameters were correlated with CSVD, indicating that OCTA is a potential method for CSVD screening.

Hypertension-related diseases have widespread effects on the systemic microvasculature, with particularly significant impacts on the retinal vascular system. As a non-invasive window to observe vascular abnormalities, fundus imaging plays an important role in the diagnosis and prediction of hypertension-related conditions. In recent years, deep learning (DL) has rapidly advanced in the field of color fundus photography (CFP) analysis, demonstrating strong potential in vessel segmentation, artery/vein classification, lesion detection, and systemic disease prediction. This review systematically summarizes recent progress in DL-based fundus analysis for hypertension-related diseases, focusing on hypertensive retinopathy analysis, automated diagnosis of ocular conditions, and cardiovascular risk prediction. Studies have shown that DL can accurately extract retinal vascular structures and pathological features, offering reliable support for early screening and risk stratification of hypertension-related diseases. Nonetheless, current models still face challenges in generalizability, robustness to low-quality images, and clinical interpretability. Future research should emphasize multimodal data integration, lightweight model design, and clinical validation to promote the real-world application of these technologies in the management of hypertension-related diseases.

Purpose: This study aimed to investigate the effects of O-linked N-acetylglucosamine modification (O-GlcNAcylation) on connexin43 (Cx43) expression and its subsequent effects on tight junction properties in diabetic retinopathy (DR). Methods: O-GlcNAcylation levels in primary human retinal vascular endothelial cells (HRVECs) and retinas from rats with diabetes were regulated by treatment with Thiamet G or alloxan. Immunoprecipitation was used to examine the relationship between O-GlcNAcylation and Cx43 expression. Stable overexpression and knockdown of Cx43 in HRVECs were achieved using lentivirus constructs; further, their effects on occludin and zonula occluden-1 (ZO-1) expression and tight junction barrier function were determined. Results: O-GlcNAcylation level increased significantly, whereas Cx43 expression decreased in retinas from rats with diabetes and HRVECs cultured under high-glucose conditions. Immunoprecipitation revealed that Cx43 was modified by O-GlcNAcylation and phosphorylation simultaneously. O-GlcNAcylation inhibition negatively regulated both total Cx43 and phosphorylated Cx43 expression, subsequently disrupting tight junction properties. Conversely, Cx43 overexpression reversed the disruption of tight junction properties and downregulated vascular endothelial growth factor expression. Consistently, Cx43 overexpression increased transendothelial electrical resistance values in HRVEC layers. Conclusions: O-GlcNAcylation negatively regulated Cx43 expression, contributing to the disruption of the blood retinal barrier. However, O-GlcNAcylation inhibition and Cx43 overexpression could reverse the tight junction disruption. Therefore, O-GlcNAcylation inhibition is a potential target for avoiding tight junction disruption through the Cx43 pathway in DR.

#933601 - Distributions of Radial Peripapillary Capillary Density and Correlations with Retinal Nerve Fiber Layer Thickness in Normal Subjects - Full View