Metraipolis › archive › dc_a95dcf9bd00d

Explainable Diabetic Retinopathy Classification Using Vision Foundation Models — arXiv:2608.28207

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.28207
License (as archived)
CC0-1.0
Archived
2026-10-17T11:06:47.702Z
SHA-256 of the archived text
fd46393c6a2bc04ce823918a87f093d1ec0920b544ee7e6aa94a752740188b8c
Stable URL of this page
https://metraipolis.com/documento/dc_a95dcf9bd00d

How to cite

Explainable Diabetic Retinopathy Classification Using Vision Foundation Models — arXiv:2608.28207 — arXiv (en), https://arxiv.org/abs/2608.28207 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a95dcf9bd00d (Metraipolis archive dc_a95dcf9bd00d, sha256 fd46393c6a2bc04ce823918a87f093d1ec0920b544ee7e6aa94a752740188b8c)
BibTeX
@misc{dca95dcf9bd00d,
  title = {Explainable Diabetic Retinopathy Classification Using Vision Foundation Models — arXiv:2608.28207},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.28207},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a95dcf9bd00d; sha256 fd46393c6a2bc04ce823918a87f093d1ec0920b544ee7e6aa94a752740188b8c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.28207. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Diabetic retinopathy (DR) is a major cause of preventable blindness, creating a need for accurate and trustworthy automated screening. This study investigates an explainable DR classification framework using vision foundation models and multiple transfer learning strategies. Three backbones, DINOv2, CLIP, and Vision Transformer (ViT), were evaluated using 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 of 0.758, while DINOv2 full fine-tuning and ViT full fine-tuning achieved the highest external AUROC of 0.920. Calibration was further assessed using reliability analysis after isotonic regression. For explainability, Grad-CAM and HiResCAM were evaluated against expert-annotated lesion masks from the IDRiD dataset using Dice, Intersection over Union (IoU), and Pointing Game metrics. The results demonstrate that foundation models, particularly DINOv2, can provide strong predictive performance, while LoRA offers a parameter-efficient alternative to full fine-tuning. Quantitative evaluation of explanation maps further supports the assessment of whether model attention corresponds to clinically relevant retinal lesions.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-31 in cs.LG).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.28207 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T10:50:25.849Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).