Benchmarking diffusion models against state-of-the-art architectures for OCT fluid biomarker segmentation
This study demonstrates that while diffusion models can comparably segment retinal pathologies using a limited number of manually annotated scans, the nnU-Net model remains the most effective overall for automated OCT analysis.
Authors
- Kiran Kumar Vupparaboina (ORCID: https://orcid.org/0000-0003-0024-8404)
- Katherine Du (ORCID: https://orcid.org/0000-0002-1715-5426)
- Adarsh Gadari (ORCID: https://orcid.org/0009-0008-8238-1957)
- José‐Alain Sahel (ORCID: https://orcid.org/0000-0002-4831-1153)
- Sandeep Chandra Bollepalli (ORCID: https://orcid.org/0000-0002-9971-9335)
- Elham Sadeghi (ORCID: https://orcid.org/0000-0003-3802-3219)
- Eveline Y. Wu (ORCID: https://orcid.org/0000-0002-6394-6843)
- Jessica Jiang (ORCID: https://orcid.org/0009-0000-7705-6211)
- Sharat Chandra Vupparaboina (ORCID: https://orcid.org/0009-0003-5957-4316)
- Utkarsh Doshi (ORCID: https://orcid.org/0009-0006-5637-3772)
- Benjamin DiCenzo
Institutions
- University of Pittsburgh Medical Center (US)
Publication Details
- Journal
- PLoS ONE
- Published
- 2025-10-29
- DOI
- https://doi.org/10.1371/journal.pone.0335615
- Primary Topic
- Optical Coherence Tomography Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00