Glaucoma is a leading eye disease, causing vision loss by gradually affecting peripheral vision if left untreated. Current diagnosis of glaucoma is performed by ophthalmologists, human experts who typically need to analyze different types of medical images generated by different types of medical equipment: fundus, Retinal Nerve Fiber Layer (RNFL), Optical Coherence Tomography (OCT) disc, OCT macula, perimetry, and/or perimetry deviation. Capturing and analyzing these medical images is labor intensive and time consuming. In this paper, we present a novel approach for glaucoma diagnosis and localization, only relying on fundus images that are analyzed by making use of state-of-the-art deep learning techniques. Specifically, our approach towards glaucoma diagnosis and localization leverages Convolutional Neural Networks (CNNs) and Gradient-weighted Class Activation Mapping (Grad-CAM), respectively. We built and evaluated different predictive models using a large set of fundus images, collected and labeled by ophthalmologists at Samsung Medical Center (SMC). Our experimental results demonstrate that our most effective predictive model is able to achieve a high diagnosis accuracy of 96%, as well as a high sensitivity of 96% and a high specificity of 100% for Dataset-Optic Disc (OD), a set of center-cropped fundus images highlighting the optic disc. Furthermore, we present Medinoid, a publicly-available prototype web application for computer-aided diagnosis and localization of glaucoma, integrating our most effective predictive model in its back-end.

This study aimed to develop and validate a deep learning system for diagnosing glaucoma using optical coherence tomography (OCT). A training set of 1822 eyes (332 control, 1490 glaucoma) with 7288 OCT images, an internal validation set of 425 eyes (104 control, 321 glaucoma) with 1700 images, and an external validation set of 355 eyes (108 control, 247 glaucoma) with 1420 images were included. Deviation and thickness maps of retinal nerve fiber layer (RNFL) and ganglion cell–inner plexiform layer (GCIPL) analyses were used to develop the deep learning system for glaucoma diagnosis based on the visual geometry group deep convolutional neural network (VGG-19) model. The diagnostic abilities of deep learning models using different OCT maps were evaluated, and the best model was compared with the diagnostic results produced by two glaucoma specialists. The glaucoma-diagnostic ability was highest when the deep learning system used the RNFL thickness map alone (area under the receiver operating characteristic curve (AUROC) 0.987), followed by the RNFL deviation map (AUROC 0.974), the GCIPL thickness map (AUROC 0.966), and the GCIPL deviation map (AUROC 0.903). Among combination sets, use of the RNFL and GCIPL deviation map showed the highest diagnostic ability, showing similar results when tested via an external validation dataset. The inclusion of the axial length did not significantly affect the diagnostic performance of the deep learning system. The location of glaucomatous damage showed generally high level of agreement between the heatmap and the diagnosis of glaucoma specialists, with 90.0% agreement when using the RNFL thickness map and 88.0% when using the GCIPL thickness map. In conclusion, our deep learning system showed high glaucoma-diagnostic abilities using OCT thickness and deviation maps. It also showed detection patterns similar to those of glaucoma specialists, showing promising results for future clinical application as an interpretable computer-aided diagnosis.

To evaluate the long-term outcomes of Argon laser photocoagulation compared to surgical direct cyclopexy in small-size cyclodialysis cleft patients. This is a retrospective study. Small-size cyclodialysis cleft patients who underwent Argon laser photocoagulation and surgical direct cyclopexy were reviewed. The mean follow-up period were 82.4 (range, 61 – 145) months and 99.9 (range, 62 – 184) months in both groups. The comparison of best corrected visual acuity (BCVA), intraocular pressure (IOP), postoperative peak IOP and time to normalization of IOP before and after the treatment. The causes of all included 15 cyclodialysis cleft cases were blunt trauma. seven patients underwent Argon laser photocoagulation and eight patients underwent surgical direct cyclopexy. The mean age of included patients was not significantly different (p = 0.38). Preoperatively, the mean logMAR BCVA (standard deviation, SD) was 0.7 (0.2) and 1.1 (0.9) and mean IOP was 4.4 (2.4) mmHg and 3.0 (1.5) mmHg in Argon laser group and surgical direct cyclopexy group (p = 0.24 and p = 0.18, respectively). The extension of cyclodialysis and duration of cyclodialysis cleft were not significantly different between the two groups (p = 0.08 and p = 0.24, respectively). The mean follow-up period were 82.4 (range, 61 – 145) months and 99.9 (range, 62 – 184) months in both groups (p = 0.41). Postoperatively, the mean logMAR BCVA was 0.0 (0.1) and 0.2 (0.3) and mean IOP was 14.5 (3.1) mmHg and 16.8 (2.5) mmHg (p = 0.15 and p = 0.16, respectively). Postoperative peak IOP and time to normalization of IOP were not different between the two groups (p = 0.75 and p = 0.91, respectively). It is necessary to use invasive treatment such as cryotherapy or surgical direct cyclopexy in cyclodialysis cleft with hypotonic maculopathy. In the present study, Argon laser photocoagulation showed good prognosis in a small-size cyclodialysis cleft below 1.5 clock-hours. Considering possible complications and cost of surgical direct cyclopexy, Argon laser can be more beneficial than surgical direct cyclopexy in small-size cyclodialysis cleft below 1.5 clock-hours. The clinical ourcome of Argon laser photocoagulation seems to be as good as surgical direct cyclopexy in small-size cyclodialysis cleft below 1.5 clock-hours.

Various substances within the aqueous humor (AH) can directly or indirectly impact intraocular tissues associated with intraocular pressure (IOP), a critical factor in glaucoma development. This study aims to investigate individual changes in these AH substances and the interactions among altered components through a multi‐omics approach. LC/MS analysis was conducted on AH samples from patients with exfoliation syndrome (XFS, n = 5), exfoliation glaucoma (XFG, n = 4), primary open‐angle glaucoma (POAG, n = 11), and cataracts (control group, n = 7). Subsequently, differentially expressed proteins and metabolites among groups, alterations in their network interactions, and their biological functions were examined. Both data‐independent acquisition and data‐dependent acquisition methods were employed to analyze the AH proteome and metabolome, and the results were integrated for a comprehensive analysis. In the proteomics analysis, proteins upregulated in both the XFG and POAG groups were associated with lipid metabolism, complement activation, and extracellular matrix regulation. Metabolomic analysis highlighted significant changes in amino acids related to antioxidant processes in the glaucoma groups. Notably, VTN, APOA1, C6, and L‐phenylalanine exhibited significant alterations in the glaucoma groups. Integration of individual omics analyses demonstrated that substances associated with inflammation and lipid metabolism, altered in the glaucoma groups, showed robust interactions within a complex network involving PLG, APOA1, and L‐phenylalanine or C3, APOD, and L‐valine. These findings offer valuable insights into the molecular mechanisms governing IOP regulation and may contribute to the development of new biomarkers for managing glaucoma.