Ophthalmology foundation models for clinically significant age macular degeneration detection

Abstract Objective . Self-supervised learning (SSL) has enabled vision transformers (ViTs) to learn robust representations from large-scale natural image datasets, enhancing their generalization across domains. In retinal imaging, foundation models pretrained on either natural or ophthalmic data have shown promise, but the benefits of in-domain pretraining remain uncertain. Approach . To investigate this, we benchmark six SSL-pretrained ViTs on seven digital fundus image (DFI) datasets totaling 70 000 expert-annotated images for the task of moderate-to-late age-related macular degeneration (AMD) identification. Main results . Our results show that DINOv2, pretrained on natural images, shows similar performance than domain-specific models. These findings highlight the value of foundation models in improving AMD identification, and challenge the assumption that in-domain pretraining is necessary. Significance . We present our model AMDNet, which performs state-of-the-art out-of-domain AUROCs on six public datasets. Furthermore, we release BRAMD, an open-access dataset ( n = 587) of DFIs with AMD labels from Brazil. Project page: www.aimlab-technion.com/lirot-ai .