Abstract

Purpose: Near-infrared reflectance imaging is a widely available but underused modality for assessing geographic atrophy (GA), a late-stage manifestation of dry age-related macular degeneration. The aim of this study was to develop and evaluate a fully automated deep-learning-based approach for detecting GA on near-infrared reflectance imaging.

Methods: Near-infrared reflectance images of patients aged 50 years or older with GA, confirmed by two retinal specialists, were analyzed at Rabin Medical Center. The control group included near-infrared reflectance images of patients with healthy-appearing retinas. Models were trained and evaluated based on accuracy, precision, sensitivity, F1 Score, and DICE coefficient.

Results: A total of 113 GA patients and 119 controls were included. The classification data set contained 330 images, and the localization data set included 659 images. Classification models performed well, with accuracy above 95%, while Vision Transformer B16 achieved the best results (precision = 98.5%, sensitivity = 98.4% and accuracy = 98.5%). For GA localization, YOLOv8-Large achieved 91% sensitivity, 91% precision, IoU of 84%, and DICE coefficient of 88%.

Conclusion: Geographic atrophy can be reliably identified using near-infrared reflectance images. Deep learning models can assist in evaluating GA on this routinely available imaging modality, aiding in the selection of patients who may benefit from emerging therapies.