Predicting branch retinal vein occlusion development using multimodal deep learning and pre-onset fundus hemisection images
Abstract Branch retinal vein occlusion (BRVO) is a leading cause of visual impairment in working-age individuals, though predicting its occurrence from retinal vascular features alone remains challenging. We developed a deep learning model to predict BRVO based on pre-onset, metadata-matched fundus...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-85777-7 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832585883228831744 |
|---|---|
| author | Eun Young Choi Dongyoung Kim Jinyeong Kim Eunjin Kim Hyunseo Lee Jinyoung Yeo Tae Keun Yoo Min Kim |
| author_facet | Eun Young Choi Dongyoung Kim Jinyeong Kim Eunjin Kim Hyunseo Lee Jinyoung Yeo Tae Keun Yoo Min Kim |
| author_sort | Eun Young Choi |
| collection | DOAJ |
| description | Abstract Branch retinal vein occlusion (BRVO) is a leading cause of visual impairment in working-age individuals, though predicting its occurrence from retinal vascular features alone remains challenging. We developed a deep learning model to predict BRVO based on pre-onset, metadata-matched fundus hemisection images. This retrospective cohort study included patients diagnosed with unilateral BRVO from two Korean tertiary centers (2005–2023), using hemisection fundus images from 27 BRVO-affected eyes paired with 81 unaffected hemisections (27 counter and 54 contralateral) for training. A U-net model segmented retinal optic discs and blood vessels (BVs), dividing them into upper and lower halves labeled for BRVO occurrence. Both unimodal models (using either fundus or BV images) and a BV-enhanced multimodal model were constructed to predict future BRVO. The multimodal model outperformed the unimodal models achieving an area under the receiver operating characteristic curve of 0.76 (95% confidence interval [CI], 0.66–0.83) and accuracy of 68.5% (95% CI 58.9–77.1%), with predictions focusing on arteriovenous crossing regions in the retinal vascular arcade. These findings demonstrate the potential of the BV-enhanced multimodal approach for BRVO prediction and highlight the need for larger, multicenter datasets to improve its clinical utility and predictive accuracy. |
| format | Article |
| id | doaj-art-e8aeae7b54444ab4a440a3fb73526bf7 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-e8aeae7b54444ab4a440a3fb73526bf72025-01-26T12:24:39ZengNature PortfolioScientific Reports2045-23222025-01-0115111210.1038/s41598-025-85777-7Predicting branch retinal vein occlusion development using multimodal deep learning and pre-onset fundus hemisection imagesEun Young Choi0Dongyoung Kim1Jinyeong Kim2Eunjin Kim3Hyunseo Lee4Jinyoung Yeo5Tae Keun Yoo6Min Kim7Department of Ophthalmology, Gangnam Severance Hospital, Institute of Vision Research, Yonsei University College of MedicineVISUWORKSDepartment of Ophthalmology, Severance Eye Hospital, Institute of Vision Research, Yonsei University College of MedicineYonsei University College of MedicineYonsei University College of MedicineDepartment of Artificial Intelligence, Yonsei University College of ComputingDepartment of Ophthalmology, Hangil Eye HospitalDepartment of Ophthalmology, Gangnam Severance Hospital, Institute of Vision Research, Yonsei University College of MedicineAbstract Branch retinal vein occlusion (BRVO) is a leading cause of visual impairment in working-age individuals, though predicting its occurrence from retinal vascular features alone remains challenging. We developed a deep learning model to predict BRVO based on pre-onset, metadata-matched fundus hemisection images. This retrospective cohort study included patients diagnosed with unilateral BRVO from two Korean tertiary centers (2005–2023), using hemisection fundus images from 27 BRVO-affected eyes paired with 81 unaffected hemisections (27 counter and 54 contralateral) for training. A U-net model segmented retinal optic discs and blood vessels (BVs), dividing them into upper and lower halves labeled for BRVO occurrence. Both unimodal models (using either fundus or BV images) and a BV-enhanced multimodal model were constructed to predict future BRVO. The multimodal model outperformed the unimodal models achieving an area under the receiver operating characteristic curve of 0.76 (95% confidence interval [CI], 0.66–0.83) and accuracy of 68.5% (95% CI 58.9–77.1%), with predictions focusing on arteriovenous crossing regions in the retinal vascular arcade. These findings demonstrate the potential of the BV-enhanced multimodal approach for BRVO prediction and highlight the need for larger, multicenter datasets to improve its clinical utility and predictive accuracy.https://doi.org/10.1038/s41598-025-85777-7Branch retinal vein occlusionDeep learningMultimodal prediction modelFundus hemisection imagesRetinal vascular features |
| spellingShingle | Eun Young Choi Dongyoung Kim Jinyeong Kim Eunjin Kim Hyunseo Lee Jinyoung Yeo Tae Keun Yoo Min Kim Predicting branch retinal vein occlusion development using multimodal deep learning and pre-onset fundus hemisection images Scientific Reports Branch retinal vein occlusion Deep learning Multimodal prediction model Fundus hemisection images Retinal vascular features |
| title | Predicting branch retinal vein occlusion development using multimodal deep learning and pre-onset fundus hemisection images |
| title_full | Predicting branch retinal vein occlusion development using multimodal deep learning and pre-onset fundus hemisection images |
| title_fullStr | Predicting branch retinal vein occlusion development using multimodal deep learning and pre-onset fundus hemisection images |
| title_full_unstemmed | Predicting branch retinal vein occlusion development using multimodal deep learning and pre-onset fundus hemisection images |
| title_short | Predicting branch retinal vein occlusion development using multimodal deep learning and pre-onset fundus hemisection images |
| title_sort | predicting branch retinal vein occlusion development using multimodal deep learning and pre onset fundus hemisection images |
| topic | Branch retinal vein occlusion Deep learning Multimodal prediction model Fundus hemisection images Retinal vascular features |
| url | https://doi.org/10.1038/s41598-025-85777-7 |
| work_keys_str_mv | AT eunyoungchoi predictingbranchretinalveinocclusiondevelopmentusingmultimodaldeeplearningandpreonsetfundushemisectionimages AT dongyoungkim predictingbranchretinalveinocclusiondevelopmentusingmultimodaldeeplearningandpreonsetfundushemisectionimages AT jinyeongkim predictingbranchretinalveinocclusiondevelopmentusingmultimodaldeeplearningandpreonsetfundushemisectionimages AT eunjinkim predictingbranchretinalveinocclusiondevelopmentusingmultimodaldeeplearningandpreonsetfundushemisectionimages AT hyunseolee predictingbranchretinalveinocclusiondevelopmentusingmultimodaldeeplearningandpreonsetfundushemisectionimages AT jinyoungyeo predictingbranchretinalveinocclusiondevelopmentusingmultimodaldeeplearningandpreonsetfundushemisectionimages AT taekeunyoo predictingbranchretinalveinocclusiondevelopmentusingmultimodaldeeplearningandpreonsetfundushemisectionimages AT minkim predictingbranchretinalveinocclusiondevelopmentusingmultimodaldeeplearningandpreonsetfundushemisectionimages |