Shen Lin1, Zhigang Li2, Bowen Fu2, Sipeng Chen3, Xi Li1, Yang Wang4, Xiaoyi Wang1, Bin Lv1,5, Bo Xu1,6, Xiantao Song7, Yao-Jun Zhang8, Xiang Cheng9, Weijian Huang10, Jun Pu11, Qi Zhang12, Yunlong Xia13, Bai Du14, Xiangyang Ji2, Zhe Zheng1,15,16. 1. National Clinical Research Center of Cardiovascular Diseases, State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, No. 167 Beilishi Road, Xicheng District, Beijing 100037, People's Republic of China. 2. Department of Automation, Tsinghua University, Main building, Haidian District, Beijing 100084, People's Republic of China. 3. Department of Information Center, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, No. 167 Beilishi Road, Xicheng District, Beijing 100037, People's Republic of China. 4. Medical Research & Biometrics Center, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, No. 167 Beilishi Road, Xicheng District, Beijing 100037, People's Republic of China. 5. Department of Radiology, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, No. 167 Beilishi Road, Xicheng District, Beijing 100037, People's Republic of China. 6. Department of Cardiology, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, No. 167 Beilishi Road, Xicheng District, Beijing 100037, People's Republic of China. 7. Department of Cardiology, Beijing Anzhen Hospital, Capital Medical University, No. 2 Anzhen Road, Chaoyang District, Beijing 100029, People's Republic of China. 8. Department of Cardiology, Xuzhou Third People's Hospital, Xuzhou Medical University, No. 131 Huancheng Road, Huaihai Economy District, Xuzhou 221000, People's Republic of China. 9. Department of Cardiology, Wuhan Union Hospital, No. 1277 Jiefang Avenue, Jianghan District, Wuhan 430022, Hubei, People's Republic of China. 10. Department of Cardiology, The First Affiliated Hospital of Wenzhou Medical University, Nanbaixiang Road, Ouhai District, Wenzhou 325000, People's Republic of China. 11. Department of Cardiology, RenJi Hospital, Shanghai JiaoTong University Medical College, No. 160 Pujian Road, Pudong New District, Shanghai 200120, People's Republic of China. 12. Department of Cardiology, Shanghai East Hospital, Tongji University School of Medicine, No. 150 Jimo Road, Pudong New District, Shanghai 200120, People's Republic of China. 13. Department of Cardiology, The First Affiliated Hospital of Dalian Medical University, No. 222 Zhongshan Road, Xigang District, Dalian 116011, People's Republic of China. 14. Department of Cardiology, Guang'anmen Hospital, China Academy of Chinese Medical Sciences, No.5 Beixiange Road, Xicheng District, Beijing 100053, People's Republic of China. 15. Department of Cardiovascular Surgery, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, No. 167 Beilishi Road, Xicheng District, Beijing 100037, People's Republic of China. 16. National Health Commission Key Laboratory of Cardiovascular Regenerative Medicine, Fuwai Central-China Hospital, Central-China Branch of National Center for Cardiovascular Diseases, No.1 Fuwai Avenue, Zhengdong New District, Zhengzhou 451464, People's Republic of China.
Abstract
AIMS: Facial features were associated with increased risk of coronary artery disease (CAD). We developed and validated a deep learning algorithm for detecting CAD based on facial photos. METHODS AND RESULTS: We conducted a multicentre cross-sectional study of patients undergoing coronary angiography or computed tomography angiography at nine Chinese sites to train and validate a deep convolutional neural network for the detection of CAD (at least one ≥50% stenosis) from patient facial photos. Between July 2017 and March 2019, 5796 patients from eight sites were consecutively enrolled and randomly divided into training (90%, n = 5216) and validation (10%, n = 580) groups for algorithm development. Between April 2019 and July 2019, 1013 patients from nine sites were enrolled in test group for algorithm test. Sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were calculated using radiologist diagnosis as the reference standard. Using an operating cut point with high sensitivity, the CAD detection algorithm had sensitivity of 0.80 and specificity of 0.54 in the test group; the AUC was 0.730 (95% confidence interval, 0.699-0.761). The AUC for the algorithm was higher than that for the Diamond-Forrester model (0.730 vs. 0.623, P < 0.001) and the CAD consortium clinical score (0.730 vs. 0.652, P < 0.001). CONCLUSION: Our results suggested that a deep learning algorithm based on facial photos can assist in CAD detection in this Chinese cohort. This technique may hold promise for pre-test CAD probability assessment in outpatient clinics or CAD screening in community. Further studies to develop a clinical available tool are warranted. Published on behalf of the European Society of Cardiology. All rights reserved.
AIMS: Facial features were associated with increased risk of coronary artery disease (CAD). We developed and validated a deep learning algorithm for detecting CAD based on facial photos. METHODS AND RESULTS: We conducted a multicentre cross-sectional study of patients undergoing coronary angiography or computed tomography angiography at nine Chinese sites to train and validate a deep convolutional neural network for the detection of CAD (at least one ≥50% stenosis) from patient facial photos. Between July 2017 and March 2019, 5796 patients from eight sites were consecutively enrolled and randomly divided into training (90%, n = 5216) and validation (10%, n = 580) groups for algorithm development. Between April 2019 and July 2019, 1013 patients from nine sites were enrolled in test group for algorithm test. Sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were calculated using radiologist diagnosis as the reference standard. Using an operating cut point with high sensitivity, the CAD detection algorithm had sensitivity of 0.80 and specificity of 0.54 in the test group; the AUC was 0.730 (95% confidence interval, 0.699-0.761). The AUC for the algorithm was higher than that for the Diamond-Forrester model (0.730 vs. 0.623, P < 0.001) and the CAD consortium clinical score (0.730 vs. 0.652, P < 0.001). CONCLUSION: Our results suggested that a deep learning algorithm based on facial photos can assist in CAD detection in this Chinese cohort. This technique may hold promise for pre-test CAD probability assessment in outpatient clinics or CAD screening in community. Further studies to develop a clinical available tool are warranted. Published on behalf of the European Society of Cardiology. All rights reserved.
Authors: Elad Maor; Nir Tsur; Galia Barkai; Ido Meister; Shmuel Makmel; Eli Friedman; Daniel Aronovich; Dana Mevorach; Amir Lerman; Eyal Zimlichman; Gideon Bachar Journal: Mayo Clin Proc Innov Qual Outcomes Date: 2021-05-14