Original Article
Evaluation of deep learning network model based on attention mechanism in diabetic retinopathy screening in China
Du Ziwei, Liu Jiang, Hu Hao, Yang Nan, Wen Liang, Wang Fenghua, Wang Bei, Yuan Yang, Sun Zilin
Published 2021-12-27
Cite as Chin J Diabetes Mellitus, 2021, 13(12): 1148-1154. DOI: 10.3760/cma.j.cn115791-20210314-00152
Abstract
ObjectiveTo evaluate the diagnostic accuracy of the SG-DR screening system (the ophthalmic image intelligent recognition software designed and produced by Zhuhai Shang Gong Co., China. Guangdong Registration Certificate for Medical Device No. 20172700901) in detecting diabetic retinopathy (DR) in the general and diabetes population, using single-or double-field fundus photography.
MethodsThe cross-sectional database was used to analyze among a randomized cluster stratification sample of 18-70 years old residents in six different ethnic groups in eight provinces of China from December 2016 to June 2017. For 8 948 enrolled participants, two retinal fundi (one disc centered 45° and one macula centered 45°) images per eye were collected for the evaluation of DR. All 17 118 images were graded by the Early Treatment Diabetic Retinopathy Study (ETDRS) scale (gold standard for DR grading), and the SG-DR screening system. Area under the receiver operating characteristic curve (AUC), sensitivity and specificity of the SG-DR screening system were performed by the receiver operating characteristic (ROC) curve based on the DR in need of referral (ETDRS>31).
ResultsIn the general population, the AUC for screening DR in need of referral using SG-DR single-field fundus photography screening system was 0.941 with sensitivity of 98.15% and specificity of 90.08% respectively. Using SG-DR double-field fundus photography screening, the sensitivity and specificity were 100% and 86.91%, respectively. In the diabetes group, for DR in need of referral, the AUC, sensitivity and specificity were 0.901, 98.08%, and 82.10%, respectively.
ConclusionsThe SG-DR screening system has shown high sensitivity and specificity in detecting DR both in the general population and diabetic group and could be as an adjunct in the screening of DR.
Key words:
Diabetic retinopathy; Artificial intelligence; Screening
Contributor Information
Du Ziwei
Department of Endocrinology, Zhongda Hospital, Institute of Diabetes, School of Medicine, Southeast University, Nanjing 210009, China
Liu Jiang
Department of Endocrinology, Zhongda Hospital, Institute of Diabetes, School of Medicine, Southeast University, Nanjing 210009, China
Hu Hao
Department of Endocrinology, Zhongda Hospital, Institute of Diabetes, School of Medicine, Southeast University, Nanjing 210009, China
Yang Nan
Department of Ophthalmology, Zhongda Hospital, School of Medicine, Southeast University, Nanjing 210009, China
Wen Liang
Fushun Eye Hospital, Fushun 113006, China
Wang Fenghua
Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing Ophthalmology &
Visual Science Key Lab, Beijing 100176, China
Wang Bei
Department of Epidemiology and Statistics, School of Public Health, Southeast University, Nanjing 210009, China
Yuan Yang
Department of Endocrinology, Zhongda Hospital, Institute of Diabetes, School of Medicine, Southeast University, Nanjing 210009, China
Sun Zilin
Department of Endocrinology, Zhongda Hospital, Institute of Diabetes, School of Medicine, Southeast University, Nanjing 210009, China
