Multimodal deep learning of fundus abnormalities and traditional risk factors for cardiovascular risk prediction
Abstract Cardiovascular disease (CVD), the leading cause of death globally, is associated with complicated underlying risk factors. We develop an artificial intelligence model to identify CVD using multimodal data, including clinical risk factors and fundus photographs from the Samsung Medical Cente...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-02-01
|
| Series: | npj Digital Medicine |
| Online Access: | https://doi.org/10.1038/s41746-023-00748-4 |
| _version_ | 1827611823411560448 |
|---|---|
| author | Yeong Chan Lee Jiho Cha Injeong Shim Woong-Yang Park Se Woong Kang Dong Hui Lim Hong-Hee Won |
| author_facet | Yeong Chan Lee Jiho Cha Injeong Shim Woong-Yang Park Se Woong Kang Dong Hui Lim Hong-Hee Won |
| author_sort | Yeong Chan Lee |
| collection | DOAJ |
| description | Abstract Cardiovascular disease (CVD), the leading cause of death globally, is associated with complicated underlying risk factors. We develop an artificial intelligence model to identify CVD using multimodal data, including clinical risk factors and fundus photographs from the Samsung Medical Center (SMC) for development and internal validation and from the UK Biobank for external validation. The multimodal model achieves an area under the receiver operating characteristic curve (AUROC) of 0.781 (95% confidence interval [CI] 0.766–0.798) in the SMC and 0.872 (95% CI 0.857–0.886) in the UK Biobank. We further observe a significant association between the incidence of CVD and the predicted risk from at-risk patients in the UK Biobank (hazard ratio [HR] 6.28, 95% CI 4.72–8.34). We visualize the importance of individual features in photography and traditional risk factors. The results highlight that non-invasive fundus photography can be a possible predictive marker for CVD. |
| first_indexed | 2024-03-09T08:13:34Z |
| format | Article |
| id | doaj.art-5f1c731bf4134a53b889ede391b8cd66 |
| institution | Directory Open Access Journal |
| issn | 2398-6352 |
| language | English |
| last_indexed | 2024-03-09T08:13:34Z |
| publishDate | 2023-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Digital Medicine |
| spelling | doaj.art-5f1c731bf4134a53b889ede391b8cd662023-12-02T22:57:28ZengNature Portfolionpj Digital Medicine2398-63522023-02-016111010.1038/s41746-023-00748-4Multimodal deep learning of fundus abnormalities and traditional risk factors for cardiovascular risk predictionYeong Chan Lee0Jiho Cha1Injeong Shim2Woong-Yang Park3Se Woong Kang4Dong Hui Lim5Hong-Hee Won6Department of Digital Health, Samsung Advanced Institute for Health Sciences & Technology (SAIHST), Sungkyunkwan University, Samsung Medical CenterGraduate School of Future Strategy, Korea Advanced Institute of Science and Technology (KAIST)Department of Digital Health, Samsung Advanced Institute for Health Sciences & Technology (SAIHST), Sungkyunkwan University, Samsung Medical CenterSamsung Genome Institute, Samsung Medical Center, Sungkyunkwan University School of MedicineDepartment of Ophthalmology, Samsung Medical Center, Sungkyunkwan University School of MedicineDepartment of Digital Health, Samsung Advanced Institute for Health Sciences & Technology (SAIHST), Sungkyunkwan University, Samsung Medical CenterDepartment of Digital Health, Samsung Advanced Institute for Health Sciences & Technology (SAIHST), Sungkyunkwan University, Samsung Medical CenterAbstract Cardiovascular disease (CVD), the leading cause of death globally, is associated with complicated underlying risk factors. We develop an artificial intelligence model to identify CVD using multimodal data, including clinical risk factors and fundus photographs from the Samsung Medical Center (SMC) for development and internal validation and from the UK Biobank for external validation. The multimodal model achieves an area under the receiver operating characteristic curve (AUROC) of 0.781 (95% confidence interval [CI] 0.766–0.798) in the SMC and 0.872 (95% CI 0.857–0.886) in the UK Biobank. We further observe a significant association between the incidence of CVD and the predicted risk from at-risk patients in the UK Biobank (hazard ratio [HR] 6.28, 95% CI 4.72–8.34). We visualize the importance of individual features in photography and traditional risk factors. The results highlight that non-invasive fundus photography can be a possible predictive marker for CVD.https://doi.org/10.1038/s41746-023-00748-4 |
| spellingShingle | Yeong Chan Lee Jiho Cha Injeong Shim Woong-Yang Park Se Woong Kang Dong Hui Lim Hong-Hee Won Multimodal deep learning of fundus abnormalities and traditional risk factors for cardiovascular risk prediction npj Digital Medicine |
| title | Multimodal deep learning of fundus abnormalities and traditional risk factors for cardiovascular risk prediction |
| title_full | Multimodal deep learning of fundus abnormalities and traditional risk factors for cardiovascular risk prediction |
| title_fullStr | Multimodal deep learning of fundus abnormalities and traditional risk factors for cardiovascular risk prediction |
| title_full_unstemmed | Multimodal deep learning of fundus abnormalities and traditional risk factors for cardiovascular risk prediction |
| title_short | Multimodal deep learning of fundus abnormalities and traditional risk factors for cardiovascular risk prediction |
| title_sort | multimodal deep learning of fundus abnormalities and traditional risk factors for cardiovascular risk prediction |
| url | https://doi.org/10.1038/s41746-023-00748-4 |
| work_keys_str_mv | AT yeongchanlee multimodaldeeplearningoffundusabnormalitiesandtraditionalriskfactorsforcardiovascularriskprediction AT jihocha multimodaldeeplearningoffundusabnormalitiesandtraditionalriskfactorsforcardiovascularriskprediction AT injeongshim multimodaldeeplearningoffundusabnormalitiesandtraditionalriskfactorsforcardiovascularriskprediction AT woongyangpark multimodaldeeplearningoffundusabnormalitiesandtraditionalriskfactorsforcardiovascularriskprediction AT sewoongkang multimodaldeeplearningoffundusabnormalitiesandtraditionalriskfactorsforcardiovascularriskprediction AT donghuilim multimodaldeeplearningoffundusabnormalitiesandtraditionalriskfactorsforcardiovascularriskprediction AT hongheewon multimodaldeeplearningoffundusabnormalitiesandtraditionalriskfactorsforcardiovascularriskprediction |