Realistic fundus photograph generation for improving automated disease classification
Latent diffusion models generated highly realistic retinal images, as validated by human experts. Adding generated images to the training set improved performance of a CNN ensemble without requiring additional real patient data.
Authors
- Prashant Pandey (ORCID: https://orcid.org/0000-0001-9275-590X)
- Peter J. Kertes (ORCID: https://orcid.org/0000-0002-0348-5096)
- Kenneth T. Eng
- Jovi C. Y. Wong (ORCID: https://orcid.org/0000-0001-7980-9352)
- Stephan Ong Tone (ORCID: https://orcid.org/0000-0001-8619-1490)
- Jonathan Andrew Micieli (ORCID: https://orcid.org/0000-0003-4911-9152)
Institutions
- St. Michael's Hospital (CA)
- Sunnybrook Health Science Centre (CA)
- University of British Columbia (CA)
- University of Toronto (CA)
- Health Sciences Centre (CA)
- Kensington Health (CA)
- Sunnybrook Hospital (CA)
Publication Details
- Journal
- British Journal of Ophthalmology
- Published
- 2025-02-12
- DOI
- https://doi.org/10.1136/bjo-2024-326122
- Citations
- 2
- Primary Topic
- Retinal Imaging and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 2.72
Funders
- Foundation Fighting Blindness
- University of Toronto
- Canadian Medical Association
- Temerty Faculty of Medicine, University of Toronto