Distributed training of foundation models for ophthalmic diagnosis
Abstract Vision impairment affects nearly 2.2 billion people globally, and nearly half of these cases could be prevented with early diagnosis and intervention—underscoring the urgent need for reliable and scalable detection methods for conditions like diabetic retinopathy and age-related macular deg...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Communications Engineering |
| Online Access: | https://doi.org/10.1038/s44172-025-00341-5 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832585701103763456 |
|---|---|
| author | Sina Gholami Fatema-E Jannat Atalie Carina Thompson Sally Shin Yee Ong Jennifer I. Lim Theodore Leng Hamed Tabkhivayghan Minhaj Nur Alam |
| author_facet | Sina Gholami Fatema-E Jannat Atalie Carina Thompson Sally Shin Yee Ong Jennifer I. Lim Theodore Leng Hamed Tabkhivayghan Minhaj Nur Alam |
| author_sort | Sina Gholami |
| collection | DOAJ |
| description | Abstract Vision impairment affects nearly 2.2 billion people globally, and nearly half of these cases could be prevented with early diagnosis and intervention—underscoring the urgent need for reliable and scalable detection methods for conditions like diabetic retinopathy and age-related macular degeneration. Here we propose a distributed deep learning framework that integrates self-supervised and domain-adaptive federated learning to enhance the detection of eye diseases from optical coherence tomography images. We employed a self-supervised, mask-based pre-training strategy to develop a robust foundation encoder. This encoder was trained on seven optical coherence tomography datasets, and we compared its performance under local, centralized, and federated learning settings. Our results show that self-supervised methods—both centralized and federated—improved the area under the curve by at least 10% compared to local models. Additionally, incorporating domain adaptation into the federated learning framework further boosted performance and generalization across different populations and imaging conditions. This approach supports collaborative model development without data sharing, providing a scalable, privacy-preserving solution for effective retinal disease screening and diagnosis in diverse clinical settings. |
| format | Article |
| id | doaj-art-41fcf8ab1ed54cbfb5e12220de9de173 |
| institution | Kabale University |
| issn | 2731-3395 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Communications Engineering |
| spelling | doaj-art-41fcf8ab1ed54cbfb5e12220de9de1732025-01-26T12:36:03ZengNature PortfolioCommunications Engineering2731-33952025-01-014111310.1038/s44172-025-00341-5Distributed training of foundation models for ophthalmic diagnosisSina Gholami0Fatema-E Jannat1Atalie Carina Thompson2Sally Shin Yee Ong3Jennifer I. Lim4Theodore Leng5Hamed Tabkhivayghan6Minhaj Nur Alam7Department of Electrical and Computer Engineering, University of North Carolina at CharlotteDepartment of Electrical and Computer Engineering, University of North Carolina at CharlotteDepartment of Ophthalmology, Wake Forest School of MedicineDepartment of Ophthalmology, Wake Forest School of MedicineDepartment of Ophthalmology and Visual Science, University of ILlinois at ChicagoByers Eye Institute at Stanford, Stanford University School of MedicineDepartment of Electrical and Computer Engineering, University of North Carolina at CharlotteDepartment of Electrical and Computer Engineering, University of North Carolina at CharlotteAbstract Vision impairment affects nearly 2.2 billion people globally, and nearly half of these cases could be prevented with early diagnosis and intervention—underscoring the urgent need for reliable and scalable detection methods for conditions like diabetic retinopathy and age-related macular degeneration. Here we propose a distributed deep learning framework that integrates self-supervised and domain-adaptive federated learning to enhance the detection of eye diseases from optical coherence tomography images. We employed a self-supervised, mask-based pre-training strategy to develop a robust foundation encoder. This encoder was trained on seven optical coherence tomography datasets, and we compared its performance under local, centralized, and federated learning settings. Our results show that self-supervised methods—both centralized and federated—improved the area under the curve by at least 10% compared to local models. Additionally, incorporating domain adaptation into the federated learning framework further boosted performance and generalization across different populations and imaging conditions. This approach supports collaborative model development without data sharing, providing a scalable, privacy-preserving solution for effective retinal disease screening and diagnosis in diverse clinical settings.https://doi.org/10.1038/s44172-025-00341-5 |
| spellingShingle | Sina Gholami Fatema-E Jannat Atalie Carina Thompson Sally Shin Yee Ong Jennifer I. Lim Theodore Leng Hamed Tabkhivayghan Minhaj Nur Alam Distributed training of foundation models for ophthalmic diagnosis Communications Engineering |
| title | Distributed training of foundation models for ophthalmic diagnosis |
| title_full | Distributed training of foundation models for ophthalmic diagnosis |
| title_fullStr | Distributed training of foundation models for ophthalmic diagnosis |
| title_full_unstemmed | Distributed training of foundation models for ophthalmic diagnosis |
| title_short | Distributed training of foundation models for ophthalmic diagnosis |
| title_sort | distributed training of foundation models for ophthalmic diagnosis |
| url | https://doi.org/10.1038/s44172-025-00341-5 |
| work_keys_str_mv | AT sinagholami distributedtrainingoffoundationmodelsforophthalmicdiagnosis AT fatemaejannat distributedtrainingoffoundationmodelsforophthalmicdiagnosis AT ataliecarinathompson distributedtrainingoffoundationmodelsforophthalmicdiagnosis AT sallyshinyeeong distributedtrainingoffoundationmodelsforophthalmicdiagnosis AT jenniferilim distributedtrainingoffoundationmodelsforophthalmicdiagnosis AT theodoreleng distributedtrainingoffoundationmodelsforophthalmicdiagnosis AT hamedtabkhivayghan distributedtrainingoffoundationmodelsforophthalmicdiagnosis AT minhajnuralam distributedtrainingoffoundationmodelsforophthalmicdiagnosis |