Multimodal LLMs for retinal disease diagnosis via OCT: few-shot versus single-shot learning
Few-shot prompted multimodal LLMs show promise for clinical integration, particularly in identifying normal retinas, which could help streamline referral processes in primary care. While these models fall short of the diagnostic accuracy reported in established deep learning literature, they offer simple, effective tools for assisting in routine retinal disease diagnosis. Future research should focus on further validation and integrating clinical text data with imaging.
Authors
- Benjamin Scott Glicksberg (ORCID: https://orcid.org/0000-0003-4515-8090)
- Eyal Klang (ORCID: https://orcid.org/0000-0002-4567-3108)
- Mahmud Omar (ORCID: https://orcid.org/0009-0001-0438-0827)
- Girish N. Nadkarni (ORCID: https://orcid.org/0000-0001-6319-4314)
- Ofira Zloto (ORCID: https://orcid.org/0000-0002-3548-424X)
- Reem Agbareia (ORCID: https://orcid.org/0009-0000-8030-9232)
Institutions
- Tel Aviv University (IL)
- Hebrew University of Jerusalem (IL)
- Sheba Medical Center (IL)
- Hadassah Medical Center (IL)
- Mount Sinai Medical Center (US)
- Icahn School of Medicine at Mount Sinai (US)
Publication Details
- Journal
- Therapeutic Advances in Ophthalmology
- Published
- 2025-05-20
- DOI
- https://doi.org/10.1177/25158414251340569
- Citations
- 8
- Primary Topic
- Retinal Imaging and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 10.88