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.

We compared 1-year outcomes of trabeculectomy with mitomycin C (MMC) and Ahmed valve implantation with MMC as a first surgical procedure in patients with uveitic glaucoma. A total 38 eyes of 38 patients undergoing trabeculectomy (n =16) or Ahmed valve implantation (n = 22) were included. Surgical success was defined as intraocular pressure (IOP) ≤21 mmHg, IOP reduction ≥20% from baseline, no secondary glaucoma surgery, and no loss of light perception. The main outcome measurements including success rate, IOP, and the number of antiglaucoma medications and complications were compared. The overall success rates were comparable between the Ahmed and trabeculectomy groups (81.3 vs. 81.8%, p = 0.987). The mean IOPs were similar as well (p = 0.084), though the number of antiglaucoma medications was significantly lower in the trabeculectomy group than in the Ahmed group (1.0 ± 1.2 vs. 2.2 ± 1.1; p = 0.005). A statistically significant reduction in corneal endothelial cell density was noted in the Ahmed group (p = 0.004). Both treatments offered reasonable IOP control and safety for eyes with uveitic glaucoma. However, significantly fewer antiglaucoma medications were used in the trabeculectomy group. Furthermore, our results suggest that cautious postoperative monitoring with regard to corneal endothelial cell density should be additionally performed after Ahmed valve implantation.

To investigate significant intraocular pressure (IOP) levels associated with the risk of open-angle glaucoma (OAG) in the treatment-naïve Korean population. Participants ≥20 years of age in Korea National Health and Nutrition Examination Survey 2010–2011 were divided into two groups, those with higher and lower IOP values, compared with the reference IOP value. We compared the risk of OAG in each group using regression analyses. The IOP value that yielded the highest statistical significance was determined as an IOP significantly associated with the OAG risk. A total of 7,650 participants (7,292 control, 358 OAG) were included. The mean IOP was significantly higher in OAG group (14.4 ± 2.9 mmHg), compared to control group (13.9 ± 2.7 mmHg, P = 0.022). In association with an increased risk of OAG, the significant IOP value was 18 mmHg (Odds ratio [OR] = 1.79, 95% confidence interval [CI] 1.14–2.80, P = 0.011). Additionally, sex-difference was identified and they were 19 mmHg (OR = 2.79, 95% CI 1.27–6.16, P = 0.011) in men and 18 mmHg (OR = 2.65, 95% CI 1.32–5.33, P = 0.006) in women. The IOP values associated with significantly decreased risk of glaucoma were determined to be 14 mmHg in men (OR = 0.68, 95% CI 0.47–0.99, P = 0.042) and 16 mmHg in women (OR = 0.47, 95% CI 0.27–0.81, P = 0.007). In consideration of the risk to benefit ratio, the reference IOP level for screening or setting the target IOP for treatment could be considered different from traditional 21 mmHg in Korean population.