DeepSeek-R1 vs OpenAI o1 for Ophthalmic Diagnoses and Management Plans
DeepSeek-R1 outperformed OpenAI o1 in diagnosis and management across subspecialties while lowering operating costs, supporting the potential of open-weight, reinforcement learning-augmented LLMs as scalable and cost-saving tools for clinical decision support. Further investigations should evaluate safety guardrails and assess performance of self-hosted adaptations of DeepSeek-R1 with domain-specific ophthalmic expertise to optimize clinical utility.
Authors
- Pearse A. Keane (ORCID: https://orcid.org/0000-0002-9239-745X)
- Andrew Farah (ORCID: https://orcid.org/0009-0005-3165-4829)
- Marko M. Popovic (ORCID: https://orcid.org/0000-0002-0370-5968)
- Daniel Milad (ORCID: https://orcid.org/0000-0002-0693-3421)
- Andrew Mihalache (ORCID: https://orcid.org/0000-0002-2692-0759)
- Fares Antaki (ORCID: https://orcid.org/0000-0001-6679-7276)
- Renaud Duval (ORCID: https://orcid.org/0000-0002-3845-3318)
- David Mikhail (ORCID: https://orcid.org/0000-0003-4814-6095)
- Michael Balas (ORCID: https://orcid.org/0000-0002-5948-0331)
- Jason Milad
- Rajeev H. Muni
Institutions
- St. Michael's Hospital (CA)
- Cleveland Clinic (US)
- Moorfields Eye Hospital NHS Foundation Trust (GB)
- University of Waterloo (CA)
- University of Toronto (CA)
- Hôpital Maisonneuve-Rosemont (CA)
- Doheny Eye Institute (US)
- Centre Hospitalier de l’Université de Montréal (CA)
- Moorfields Eye Hospital (GB)
- Cleveland Eye Clinic (US)
- Unity Health Toronto
- University College London (GB)
- McGill University (CA)
- Université de Montréal (CA)
Publication Details
- Journal
- JAMA Ophthalmology
- Published
- 2025-09-04
- DOI
- https://doi.org/10.1001/jamaophthalmol.2025.2918
- Citations
- 4
- Primary Topic
- Clinical Reasoning and Diagnostic Skills
- Type
- article
- Field-Weighted Citation Impact
- 6.06