Tyler Hyungtaek Rim1, Chan Joo Lee2, Yih-Chung Tham3, Ning Cheung3, Marco Yu4, Geunyoung Lee5, Youngnam Kim5, Daniel S W Ting3, Crystal Chun Yuen Chong4, Yoon Seong Choi6, Tae Keun Yoo7, Ik Hee Ryu8, Su Jung Baik9, Young Ah Kim10, Sung Kyu Kim11, Sang-Hak Lee12, Byoung Kwon Lee13, Seok-Min Kang2, Edmund Yick Mun Wong3, Hyeon Chang Kim14, Sung Soo Kim15, Sungha Park16, Ching-Yu Cheng17, Tien Yin Wong3. 1. Singapore Eye Research Institute, Singapore National Eye Centre, Singapore; Ophthalmology and Visual Sciences Academic Clinical Program, Duke-NUS Medical School, Singapore; Department of Ophthalmology, Institute of Vision Research, Severance Hospital, Yonsei University College of Medicine, Seoul, South Korea. 2. Division of Cardiology, Severance Cardiovascular Hospital, Yonsei University College of Medicine, Seoul, South Korea. 3. Singapore Eye Research Institute, Singapore National Eye Centre, Singapore; Ophthalmology and Visual Sciences Academic Clinical Program, Duke-NUS Medical School, Singapore. 4. Singapore Eye Research Institute, Singapore National Eye Centre, Singapore. 5. Medi Whale, Seoul, South Korea. 6. Radiological Sciences Academic Clinical Programme, Duke-NUS Medical School, Singapore; Department of Radiology, Research Institute of Radiological Science, Severance Hospital, Yonsei University College of Medicine, Seoul, South Korea. 7. Department of Ophthalmology, Institute of Vision Research, Severance Hospital, Yonsei University College of Medicine, Seoul, South Korea. 8. B&VIIt Eye Center, Seoul, South Korea. 9. Healthcare Research Team, Health Promotion Center, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, South Korea. 10. Division of Healthcare Big Data, Yonsei University College of Medicine, Seoul, South Korea. 11. Philip Medical Center, Bundang, Seongnam, South Korea. 12. Division of Cardiology, Severance Cardiovascular Hospital, Yonsei University College of Medicine, Seoul, South Korea; Integrated Research Center for Cerebrovascular and Cardiovascular Diseases, Yonsei University College of Medicine, Seoul, South Korea. 13. Division of Cardiology, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, South Korea. 14. Department of Preventive Medicine, Yonsei University College of Medicine, Seoul, South Korea; Integrated Research Center for Cerebrovascular and Cardiovascular Diseases, Yonsei University College of Medicine, Seoul, South Korea. 15. Department of Ophthalmology, Institute of Vision Research, Severance Hospital, Yonsei University College of Medicine, Seoul, South Korea. Electronic address: semekim@yuhs.ac. 16. Division of Cardiology, Severance Cardiovascular Hospital, Yonsei University College of Medicine, Seoul, South Korea; Integrated Research Center for Cerebrovascular and Cardiovascular Diseases, Yonsei University College of Medicine, Seoul, South Korea. Electronic address: shpark0530@yuhs.ac. 17. Singapore Eye Research Institute, Singapore National Eye Centre, Singapore; Ophthalmology and Visual Sciences Academic Clinical Program, Duke-NUS Medical School, Singapore. Electronic address: chingyu.cheng@duke-nus.edu.sg.
Abstract
BACKGROUND: Coronary artery calcium (CAC) score is a clinically validated marker of cardiovascular disease risk. We developed and validated a novel cardiovascular risk stratification system based on deep-learning-predicted CAC from retinal photographs. METHODS: We used 216 152 retinal photographs from five datasets from South Korea, Singapore, and the UK to train and validate the algorithms. First, using one dataset from a South Korean health-screening centre, we trained a deep-learning algorithm to predict the probability of the presence of CAC (ie, deep-learning retinal CAC score, RetiCAC). We stratified RetiCAC scores into tertiles and used Cox proportional hazards models to evaluate the ability of RetiCAC to predict cardiovascular events based on external test sets from South Korea, Singapore, and the UK Biobank. We evaluated the incremental values of RetiCAC when added to the Pooled Cohort Equation (PCE) for participants in the UK Biobank. FINDINGS: RetiCAC outperformed all single clinical parameter models in predicting the presence of CAC (area under the receiver operating characteristic curve of 0·742, 95% CI 0·732-0·753). Among the 527 participants in the South Korean clinical cohort, 33 (6·3%) had cardiovascular events during the 5-year follow-up. When compared with the current CAC risk stratification (0, >0-100, and >100), the three-strata RetiCAC showed comparable prognostic performance with a concordance index of 0·71. In the Singapore population-based cohort (n=8551), 310 (3·6%) participants had fatal cardiovascular events over 10 years, and the three-strata RetiCAC was significantly associated with increased risk of fatal cardiovascular events (hazard ratio [HR] trend 1·33, 95% CI 1·04-1·71). In the UK Biobank (n=47 679), 337 (0·7%) participants had fatal cardiovascular events over 10 years. When added to the PCE, the three-strata RetiCAC improved cardiovascular risk stratification in the intermediate-risk group (HR trend 1·28, 95% CI 1·07-1·54) and borderline-risk group (1·62, 1·04-2·54), and the continuous net reclassification index was 0·261 (95% CI 0·124-0·364). INTERPRETATION: A deep learning and retinal photograph-derived CAC score is comparable to CT scan-measured CAC in predicting cardiovascular events, and improves on current risk stratification approaches for cardiovascular disease events. These data suggest retinal photograph-based deep learning has the potential to be used as an alternative measure of CAC, especially in low-resource settings. FUNDING: Yonsei University College of Medicine; Ministry of Health and Welfare, Korea Institute for Advancement of Technology, South Korea; Agency for Science, Technology, and Research; and National Medical Research Council, Singapore.
BACKGROUND: Coronary artery calcium (CAC) score is a clinically validated marker of cardiovascular disease risk. We developed and validated a novel cardiovascular risk stratification system based on deep-learning-predicted CAC from retinal photographs. METHODS: We used 216 152 retinal photographs from five datasets from South Korea, Singapore, and the UK to train and validate the algorithms. First, using one dataset from a South Korean health-screening centre, we trained a deep-learning algorithm to predict the probability of the presence of CAC (ie, deep-learning retinal CAC score, RetiCAC). We stratified RetiCAC scores into tertiles and used Cox proportional hazards models to evaluate the ability of RetiCAC to predict cardiovascular events based on external test sets from South Korea, Singapore, and the UK Biobank. We evaluated the incremental values of RetiCAC when added to the Pooled Cohort Equation (PCE) for participants in the UK Biobank. FINDINGS: RetiCAC outperformed all single clinical parameter models in predicting the presence of CAC (area under the receiver operating characteristic curve of 0·742, 95% CI 0·732-0·753). Among the 527 participants in the South Korean clinical cohort, 33 (6·3%) had cardiovascular events during the 5-year follow-up. When compared with the current CAC risk stratification (0, >0-100, and >100), the three-strata RetiCAC showed comparable prognostic performance with a concordance index of 0·71. In the Singapore population-based cohort (n=8551), 310 (3·6%) participants had fatal cardiovascular events over 10 years, and the three-strata RetiCAC was significantly associated with increased risk of fatal cardiovascular events (hazard ratio [HR] trend 1·33, 95% CI 1·04-1·71). In the UK Biobank (n=47 679), 337 (0·7%) participants had fatal cardiovascular events over 10 years. When added to the PCE, the three-strata RetiCAC improved cardiovascular risk stratification in the intermediate-risk group (HR trend 1·28, 95% CI 1·07-1·54) and borderline-risk group (1·62, 1·04-2·54), and the continuous net reclassification index was 0·261 (95% CI 0·124-0·364). INTERPRETATION: A deep learning and retinal photograph-derived CAC score is comparable to CT scan-measured CAC in predicting cardiovascular events, and improves on current risk stratification approaches for cardiovascular disease events. These data suggest retinal photograph-based deep learning has the potential to be used as an alternative measure of CAC, especially in low-resource settings. FUNDING: Yonsei University College of Medicine; Ministry of Health and Welfare, Korea Institute for Advancement of Technology, South Korea; Agency for Science, Technology, and Research; and National Medical Research Council, Singapore.
Authors: Bo Ram Kim; Tae Keun Yoo; Hong Kyu Kim; Ik Hee Ryu; Jin Kuk Kim; In Sik Lee; Jung Soo Kim; Dong-Hyeok Shin; Young-Sang Kim; Bom Taeck Kim Journal: EPMA J Date: 2022-08-08 Impact factor: 8.836
Authors: Patrik Bachtiger; Camille F Petri; Francesca E Scott; Se Ri Park; Mihir A Kelshiker; Harpreet K Sahemey; Bianca Dumea; Regine Alquero; Pritpal S Padam; Isobel R Hatrick; Alfa Ali; Maria Ribeiro; Wing-See Cheung; Nina Bual; Bushra Rana; Matthew Shun-Shin; Daniel B Kramer; Alex Fragoyannis; Daniel Keene; Carla M Plymen; Nicholas S Peters Journal: Lancet Digit Health Date: 2022-01-05
Authors: Sergio Sanchez-Martinez; Oscar Camara; Gemma Piella; Maja Cikes; Miguel Ángel González-Ballester; Marius Miron; Alfredo Vellido; Emilia Gómez; Alan G Fraser; Bart Bijnens Journal: Front Cardiovasc Med Date: 2022-01-04
Authors: Alejandra Planas; Olga Simó-Servat; Cristina Hernández; Ángel Ortiz-Zúñiga; Joan Ramón Marsal; José R Herance; Ignacio Ferreira-González; Rafael Simó Journal: J Pers Med Date: 2021-12-10
Authors: Jasjit S Suri; Mrinalini Bhagawati; Sudip Paul; Athanasios D Protogerou; Petros P Sfikakis; George D Kitas; Narendra N Khanna; Zoltan Ruzsa; Aditya M Sharma; Sanjay Saxena; Gavino Faa; John R Laird; Amer M Johri; Manudeep K Kalra; Kosmas I Paraskevas; Luca Saba Journal: Diagnostics (Basel) Date: 2022-03-16