[Submitted on 28 Aug 2026 (v1), last revised 3 Nov 2026 (this version, v3)]
Abstract:Diabetic retinopathy (DR) is a leading cause of preventable blindness, and automated screening must be both accurate and trustworthy. We study an explainable DR classification framework built on vision foundation models and three transfer learning strategies. Three backbones, DINOv2, CLIP, and Vision Transformer (ViT), were adapted by full fine-tuning, linear probing, and Low-Rank Adaptation (LoRA). Models were trained and internally evaluated on the ODIR dataset and externally evaluated on APTOS to assess generalization. DINOv2-LoRA achieved the highest internal AUROC (0.758), whereas DINOv2 and ViT full fine-tuning achieved the highest external AUROC (0.920). Probability calibration was assessed with reliability analysis after isotonic regression. For explainability, Grad-CAM and HiResCAM maps were compared with expert-annotated lesion masks from the IDRiD dataset using Dice, Intersection over Union (IoU), and Pointing Game metrics. The results show that foundation models, particularly DINOv2, provide strong predictive performance, and that LoRA is a parameter-efficient alternative to full fine-tuning. Spatial overlap between explanation maps and annotated lesions was limited (mean Dice of about 0.13 for the combined lesion mask), which highlights the value of quantitatively testing whether model attention corresponds to clinically relevant retinal lesions.
Submission history
From: Juan Miguel Lopez Alcaraz [view email]
[v1] Fri, 28 Aug 2026 11:26:43 UTC (2,668 KB)
[v2] Fri, 4 Sep 2026 20:02:44 UTC (2,674 KB)
[v3] Tue, 3 Nov 2026 16:20:52 UTC (2,679 KB)
