To develop an artificial intelligence (AI) model that predicts anti-vascular endothelial growth factor (VEGF) agent-specific anatomical treatment outcomes in neovascular age-related macular degeneration (AMD), thereby assisting clinicians in selecting the most suitable anti-VEGF agent for each patient. This retrospective study included patients diagnosed with neovascular AMD who received three loading injections of either ranibizumab or aflibercept. Training was performed using optical coherence tomography (OCT) images with an attention generative adversarial network (GAN) model. To test the performance of the AI model, the sensitivity and specificity to predict the presence of retinal fluid after treatment were calculated for the AI model, an experienced (Examiner 1), and a less experienced (Examiner 2) human examiners. A total of 1684 OCT images from 842 patients (419 treated with ranibizumab and 423 treated with aflibercept) were used as the training set. Testing was performed using images from 98 patients. In patients treated with ranibizumab, the sensitivity and specificity, respectively, were 0.615 and 0.667 for the AI model, 0.385 and 0.861 for Examiner 1, and 0.231 and 0.806 for Examiner 2. In patients treated with aflibercept, the sensitivity and specificity, respectively, were 0.857 and 0.881 for the AI model, 0.429 and 0.976 for Examiner 1, and 0.429 and 0.857 for Examiner 2. In 18.5% of cases, the fluid status of synthetic posttreatment images differed between ranibizumab and aflibercept. The AI model using GAN might predict anti-VEGF agent-specific short-term treatment outcomes with relatively higher sensitivity than human examiners. Additionally, there was a difference in the efficacy in fluid resolution between the anti-VEGF agents. These results suggest the potential of AI in personalized medicine for patients with neovascular AMD.

Purpose: To evaluate the prevalence and characteristics of subtypes of pseudodrusen in newly diagnosed exudative age-related macular degeneration (AMD) and polypoidal choroidal vasculopathy (PCV). Methods: This retrospective cross-sectional study included 321 eyes of 321 patients who were newly diagnosed with exudative AMD or PCV. Reticular pseudodrusen was classified into dot pseudodrusen and ribbon pseudodrusen; the prevalence of each subtype was estimated and compared between exudative AMD excluding retinal angiomatous proliferation (RAP), PCV, and RAP. Patient age and choroidal thickness were compared between patients with dot pseudodrusen only and those with ribbon pseudodrusen. Results: The prevalence of reticular pseudodrusen was 13.9% (15 of 108 eyes) in exudative AMD excluding RAP, 3.4% (6 of 175 eyes) in PCV, and 68.4% (27 of 38 eyes) in RAP. Among the eyes with pseudodrusen, dot pseudodrusen and ribbon pseudodrusen were noted in 100% and 40.0%, respectively, in exudative AMD excluding RAP, 100% and 16.7%, respectively, in PCV, and 96.2% and 69.2%, respectively, in RAP. Ribbon pseudodrusen was more frequently observed in RAP (P = 0.032). Patients with ribbon pseudodrusen were significantly older (77.3 ± 6.6 years vs. 72.9 ± 8.1 years, P = 0.042) than those with dot pseudodrusen only. Conclusion: The markedly higher incidence of ribbon pseudodrusen in RAP may suggest possible influence of ribbon pseudodrusen on the development of RAP.

To compare lesion reactivation and treatment outcomes between pure type 2 and mixed type 1 and 2 macular neovascularization (MNV) treated with anti-vascular endothelial growth factor (VEGF).This re...

Purpose.: To evaluate the prevalence of choroidal vascular hyperpermeability and punctate hyperfluorescent spots in eyes with choroidal neovascularization (CNV). Methods.: This retrospective, observational study included 382 eyes with typical exudative AMD (97 eyes), polypoidal choroidal vasculopathy (PCV, 163 eyes), retinal angiomatous proliferation (RAP, 37 eyes), or myopic CNV (86 eyes). The prevalence of choroidal vascular hyperpermeability and punctate hyperfluorescent spots was estimated based on available indocyanine green angiography (ICGA) images. Results.: Choroidal vascular hyperpermeability was noted in 12.4% (12 eyes) and 26.9% (42 eyes) of eyes with typical exudative AMD and PCV, respectively. Choroidal vascular hyperpermeability was not noted in any eye with RAP or myopic CNV. Punctate hyperfluorescent spots were noted in 43.3% (42 eyes), 72.4% (118 eyes), 10.8% (4 eyes), and 4.7% (4 eyes) of eyes with typical exudative AMD, PCV, RAP, and myopic CNV, respectively. Of the 56 eyes with choroidal vascular hyperpermeability, punctate hyperfluorescent spots were noted in 55 eyes (98.2%). Conclusions.: Choroidal vascular hyperpermeability and punctate hyperfluorescent spots may have a common pathophysiology. Although choroidal vascular hyperpermeability and punctate hyperfluorescent spots have been thought to be associated with pathologic conditions, the markedly low prevalence of these findings in eyes with RAP and myopic CNV may not be a normal finding. It is possible that compromised choroidal perfusion, with or without associated with choroidal thinning, may lead the low prevalence of these abnormalities in eyes with these two disorders.