Non-invasive biopsy diagnosis of diabetic kidney disease via deep learning applied to retinal images: a population-based study

Summary: Background: Improving the accessibility of screening diabetic kidney disease (DKD) and differentiating isolated diabetic nephropathy from non-diabetic kidney disease (NDKD) are two major challenges in the field of diabetes care. We aimed to develop and validate an artificial intelligence (...

Full description

Saved in:
Bibliographic Details
Main Authors: Ziyao Meng, MEng, Zhouyu Guan, MD, Shujie Yu, MBBS, Yilan Wu, BSc, Yaoning Zhao, BSc, Jie Shen, PhD, Cynthia Ciwei Lim, MMed, Tingli Chen, MD, Dawei Yang, PhD, An Ran Ran, PhD, Feng He, MSc, Haslina Hamzah, BSc, Sarkaaj Singh, MSc, Anis Syazwani Abd Raof, MSc, Jian Wen Samuel Lee-Boey, MBBS, ProfMBBS Soo-Kun Lim, MBBS, ProfMD Xufang Sun, MBBS, Shuwang Ge, MD, ProfMD Gang Xu, MD, Prof Hua Su, MD, Yang Cheng, MD, Feng Lu, PhD, ProfPhD Xiaofei Liao, PhD, ProfPhD Hai Jin, PhD, Chenxin Deng, MMed, Lei Ruan, MD, ProfMD Cuntai Zhang, MD, Chan Wu, MD, ProfMD Rongping Dai, MD, Yixiao Jin, MSc, Wenxiao Wang, PhD, Tingyao Li, BSc, Ruhan Liu, PhD, Jiajia Li, PhD, Jia Shu, BEng, Yuwei Lu, MBBS, Xiangning Wang, MD, Qiang Wu, MD, Yiming Qin, BEng, Jin Tang, MD, Xiaohua Sheng, MD, Qiong Jiao, MD, ProfPhD Xiaokang Yang, MD, ProfPhD Minyi Guo, MD, Gareth J McKay, PhD, ProfPhD Ruth E Hogg, PhD, Gerald Liew, MBBS, Evelyn Yi Lyn Chee, PhD, ProfPhD Wynne Hsu, PhD, ProfPhD Mong Li Lee, PhD, Simon Szeto, MBChB, ProfMD Andrea O Y Luk, MBChB, ProfMD Juliana C N Chan, MBChB, Carol Y Cheung, PhD, Gavin Siew Wei Tan, FRCSEd, Yih-Chung Tham, PhD, ProfMD PhD Ching-Yu Cheng, PhD, Charumathi Sabanayagam, MD PhD, ProfPhD Lee-Ling Lim, MD PhD, ProfMD PhD Weiping Jia, MD PhD, ProfMD PhD Huating Li, MD PhD, Bin Sheng, Prof, Tien Yin Wong, Prof
Format: Article
Language:English
Published: Elsevier 2025-05-01
Series:The Lancet: Digital Health
Online Access:http://www.sciencedirect.com/science/article/pii/S2589750025000408
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850183660230148096
author Ziyao Meng, MEng
Zhouyu Guan, MD
Shujie Yu, MBBS
Yilan Wu, BSc
Yaoning Zhao, BSc
Jie Shen, PhD
Cynthia Ciwei Lim, MMed
Tingli Chen, MD
Dawei Yang, PhD
An Ran Ran, PhD
Feng He, MSc
Haslina Hamzah, BSc
Sarkaaj Singh, MSc
Anis Syazwani Abd Raof, MSc
Jian Wen Samuel Lee-Boey, MBBS
ProfMBBS Soo-Kun Lim, MBBS
ProfMD Xufang Sun, MBBS
Shuwang Ge, MD
ProfMD Gang Xu, MD
Prof Hua Su, MD
Yang Cheng, MD
Feng Lu, PhD
ProfPhD Xiaofei Liao, PhD
ProfPhD Hai Jin, PhD
Chenxin Deng, MMed
Lei Ruan, MD
ProfMD Cuntai Zhang, MD
Chan Wu, MD
ProfMD Rongping Dai, MD
Yixiao Jin, MSc
Wenxiao Wang, PhD
Tingyao Li, BSc
Ruhan Liu, PhD
Jiajia Li, PhD
Jia Shu, BEng
Yuwei Lu, MBBS
Xiangning Wang, MD
Qiang Wu, MD
Yiming Qin, BEng
Jin Tang, MD
Xiaohua Sheng, MD
Qiong Jiao, MD
ProfPhD Xiaokang Yang, MD
ProfPhD Minyi Guo, MD
Gareth J McKay, PhD
ProfPhD Ruth E Hogg, PhD
Gerald Liew, MBBS
Evelyn Yi Lyn Chee, PhD
ProfPhD Wynne Hsu, PhD
ProfPhD Mong Li Lee, PhD
Simon Szeto, MBChB
ProfMD Andrea O Y Luk, MBChB
ProfMD Juliana C N Chan, MBChB
Carol Y Cheung, PhD
Gavin Siew Wei Tan, FRCSEd
Yih-Chung Tham, PhD
ProfMD PhD Ching-Yu Cheng, PhD
Charumathi Sabanayagam, MD PhD
ProfPhD Lee-Ling Lim, MD PhD
ProfMD PhD Weiping Jia, MD PhD
ProfMD PhD Huating Li, MD PhD
Bin Sheng, Prof
Tien Yin Wong, Prof
author_facet Ziyao Meng, MEng
Zhouyu Guan, MD
Shujie Yu, MBBS
Yilan Wu, BSc
Yaoning Zhao, BSc
Jie Shen, PhD
Cynthia Ciwei Lim, MMed
Tingli Chen, MD
Dawei Yang, PhD
An Ran Ran, PhD
Feng He, MSc
Haslina Hamzah, BSc
Sarkaaj Singh, MSc
Anis Syazwani Abd Raof, MSc
Jian Wen Samuel Lee-Boey, MBBS
ProfMBBS Soo-Kun Lim, MBBS
ProfMD Xufang Sun, MBBS
Shuwang Ge, MD
ProfMD Gang Xu, MD
Prof Hua Su, MD
Yang Cheng, MD
Feng Lu, PhD
ProfPhD Xiaofei Liao, PhD
ProfPhD Hai Jin, PhD
Chenxin Deng, MMed
Lei Ruan, MD
ProfMD Cuntai Zhang, MD
Chan Wu, MD
ProfMD Rongping Dai, MD
Yixiao Jin, MSc
Wenxiao Wang, PhD
Tingyao Li, BSc
Ruhan Liu, PhD
Jiajia Li, PhD
Jia Shu, BEng
Yuwei Lu, MBBS
Xiangning Wang, MD
Qiang Wu, MD
Yiming Qin, BEng
Jin Tang, MD
Xiaohua Sheng, MD
Qiong Jiao, MD
ProfPhD Xiaokang Yang, MD
ProfPhD Minyi Guo, MD
Gareth J McKay, PhD
ProfPhD Ruth E Hogg, PhD
Gerald Liew, MBBS
Evelyn Yi Lyn Chee, PhD
ProfPhD Wynne Hsu, PhD
ProfPhD Mong Li Lee, PhD
Simon Szeto, MBChB
ProfMD Andrea O Y Luk, MBChB
ProfMD Juliana C N Chan, MBChB
Carol Y Cheung, PhD
Gavin Siew Wei Tan, FRCSEd
Yih-Chung Tham, PhD
ProfMD PhD Ching-Yu Cheng, PhD
Charumathi Sabanayagam, MD PhD
ProfPhD Lee-Ling Lim, MD PhD
ProfMD PhD Weiping Jia, MD PhD
ProfMD PhD Huating Li, MD PhD
Bin Sheng, Prof
Tien Yin Wong, Prof
author_sort Ziyao Meng, MEng
collection DOAJ
description Summary: Background: Improving the accessibility of screening diabetic kidney disease (DKD) and differentiating isolated diabetic nephropathy from non-diabetic kidney disease (NDKD) are two major challenges in the field of diabetes care. We aimed to develop and validate an artificial intelligence (AI) deep learning system to detect DKD and isolated diabetic nephropathy from retinal fundus images. Methods: In this population-based study, we developed a retinal image-based AI-deep learning system, DeepDKD, pretrained using 734 084 retinal fundus images. First, for DKD detection, we used 486 312 retinal images from 121 578 participants in the Shanghai Integrated Diabetes Prevention and Care System for development and internal validation, and ten multi-ethnic datasets from China, Singapore, Malaysia, Australia, and the UK (65 406 participants) for external validation. Second, to differentiate isolated diabetic nephropathy from NDKD, we used 1068 retinal images from 267 participants for development and internal validation, and three multi-ethnic datasets from China, Malaysia, and the UK (244 participants) for external validation. Finally, we conducted two proof-of-concept studies: a prospective real-world study with 3 months' follow-up to evaluate the effectiveness of DeepDKD in screening DKD; and a longitudinal analysis of the effectiveness of DeepDKD in differentiating isolated diabetic nephropathy from NDKD on renal function changes with 4·6 years' follow-up. Findings: For detecting DKD, DeepDKD achieved an area under the receiver operating characteristic curve (AUC) of 0·842 (95% CI 0·838–0·846) on the internal validation dataset and AUCs of 0·791–0·826 across external validation datasets. For differentiating isolated diabetic nephropathy from NDKD, DeepDKD achieved an AUC of 0·906 (0·825–0·966) on the internal validation dataset and AUCs of 0·733–0·844 across external validation datasets. In the prospective study, compared with the metadata model, DeepDKD could detect DKD with higher sensitivity (89·8% vs 66·3%, p<0·0001). In the longitudinal study, participants with isolated diabetic nephropathy and participants with NDKD identified by DeepDKD had a significant difference in renal function outcomes (proportion of estimated glomerular filtration rate decline: 27·45% vs 52·56%, p=0·0010). Interpretation: Among diverse multi-ethnic populations with diabetes, a retinal image-based AI-deep learning system showed its potential for detecting DKD and differentiating isolated diabetic nephropathy from NDKD in clinical practice. Funding: National Key R &amp; D Program of China, National Natural Science Foundation of China, Beijing Natural Science Foundation, Shanghai Municipal Key Clinical Specialty, Shanghai Research Centre for Endocrine and Metabolic Diseases, Innovative research team of high-level local universities in Shanghai, Noncommunicable Chronic Diseases-National Science and Technology Major Project, Clinical Special Program of Shanghai Municipal Health Commission, and the three-year action plan to strengthen the construction of public health system in Shanghai.
format Article
id doaj-art-ae5128210fc84ef4b71df992d964e817
institution OA Journals
issn 2589-7500
language English
publishDate 2025-05-01
publisher Elsevier
record_format Article
series The Lancet: Digital Health
spelling doaj-art-ae5128210fc84ef4b71df992d964e8172025-08-20T02:17:19ZengElsevierThe Lancet: Digital Health2589-75002025-05-017510086810.1016/j.landig.2025.02.008Non-invasive biopsy diagnosis of diabetic kidney disease via deep learning applied to retinal images: a population-based studyZiyao Meng, MEng0Zhouyu Guan, MD1Shujie Yu, MBBS2Yilan Wu, BSc3Yaoning Zhao, BSc4Jie Shen, PhD5Cynthia Ciwei Lim, MMed6Tingli Chen, MD7Dawei Yang, PhD8An Ran Ran, PhD9Feng He, MSc10Haslina Hamzah, BSc11Sarkaaj Singh, MSc12Anis Syazwani Abd Raof, MSc13Jian Wen Samuel Lee-Boey, MBBS14ProfMBBS Soo-Kun Lim, MBBS15ProfMD Xufang Sun, MBBS16Shuwang Ge, MD17ProfMD Gang Xu, MD18Prof Hua Su, MD19Yang Cheng, MD20Feng Lu, PhD21ProfPhD Xiaofei Liao, PhD22ProfPhD Hai Jin, PhD23Chenxin Deng, MMed24Lei Ruan, MD25ProfMD Cuntai Zhang, MD26Chan Wu, MD27ProfMD Rongping Dai, MD28Yixiao Jin, MSc29Wenxiao Wang, PhD30Tingyao Li, BSc31Ruhan Liu, PhD32Jiajia Li, PhD33Jia Shu, BEng34Yuwei Lu, MBBS35Xiangning Wang, MD36Qiang Wu, MD37Yiming Qin, BEng38Jin Tang, MD39Xiaohua Sheng, MD40Qiong Jiao, MD41ProfPhD Xiaokang Yang, MD42ProfPhD Minyi Guo, MD43Gareth J McKay, PhD44ProfPhD Ruth E Hogg, PhD45Gerald Liew, MBBS46Evelyn Yi Lyn Chee, PhD47ProfPhD Wynne Hsu, PhD48ProfPhD Mong Li Lee, PhD49Simon Szeto, MBChB50ProfMD Andrea O Y Luk, MBChB51ProfMD Juliana C N Chan, MBChB52Carol Y Cheung, PhD53Gavin Siew Wei Tan, FRCSEd54Yih-Chung Tham, PhD55ProfMD PhD Ching-Yu Cheng, PhD56Charumathi Sabanayagam, MD PhD57ProfPhD Lee-Ling Lim, MD PhD58ProfMD PhD Weiping Jia, MD PhD59ProfMD PhD Huating Li, MD PhD60Bin Sheng, Prof61Tien Yin Wong, Prof62Shanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, China; MOE Key Laboratory of AI, School of Electronic, Information, and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaShanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, ChinaShanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, ChinaBeijing Visual Science and Translational Eye Research Institute (BERI), Beijing Tsinghua Changgung Hospital Eye Center, School of Clinical Medicine, Tsinghua Medicine, Tsinghua University, Beijing, ChinaBeijing Visual Science and Translational Eye Research Institute (BERI), Beijing Tsinghua Changgung Hospital Eye Center, School of Clinical Medicine, Tsinghua Medicine, Tsinghua University, Beijing, ChinaMedical Records and Statistics Office, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, ChinaDepartment of Renal Medicine, Singapore General Hospital, SingHealth-Duke Academic Medical Centre, Singapore, SingaporeDepartment of Ophthalmology, Shanghai Health and Medical Center, Wuxi, ChinaDepartment of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong Special Administrative Region, ChinaDepartment of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong Special Administrative Region, ChinaSingapore Eye Research Institute, Singapore National Eye Centre, SingaporeSingapore Eye Research Institute, Singapore National Eye Centre, SingaporeDepartment of Medicine, Faculty of Medicine, Universiti Malaya, Kuala Lumpur, MalaysiaDepartment of Medicine, Faculty of Medicine, Universiti Malaya, Kuala Lumpur, MalaysiaDepartment of Medicine, Faculty of Medicine, Universiti Malaya, Kuala Lumpur, MalaysiaDepartment of Medicine, Faculty of Medicine, Universiti Malaya, Kuala Lumpur, MalaysiaDepartment of Ophthalmology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei, ChinaDepartment of Nephrology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaDepartment of Nephrology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaDepartment of Nephrology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaDepartment of Ophthalmology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaNational Engineering Research Centre for Big Data Technology and System, Services Computing Technology and System Laboratory, Cluster and Grid Computing Laboratory, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, ChinaNational Engineering Research Centre for Big Data Technology and System, Services Computing Technology and System Laboratory, Cluster and Grid Computing Laboratory, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, ChinaNational Engineering Research Centre for Big Data Technology and System, Services Computing Technology and System Laboratory, Cluster and Grid Computing Laboratory, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, ChinaDepartment of Geriatrics, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei, China; Key Laboratory of Vascular Aging, Ministry of Education, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaDepartment of Geriatrics, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei, China; Key Laboratory of Vascular Aging, Ministry of Education, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaDepartment of Geriatrics, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei, China; Key Laboratory of Vascular Aging, Ministry of Education, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaDepartment of Ophthalmology, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing, ChinaDepartment of Ophthalmology, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing, ChinaBeijing Visual Science and Translational Eye Research Institute (BERI), Beijing Tsinghua Changgung Hospital Eye Center, School of Clinical Medicine, Tsinghua Medicine, Tsinghua University, Beijing, ChinaKey Laboratory of River Basic Digital Twinning of Ministry of Water Resources, Macau University of Science and Technology, Macau Special Administrative Region, ChinaShanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, China; MOE Key Laboratory of AI, School of Electronic, Information, and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaShanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, China; MOE Key Laboratory of AI, School of Electronic, Information, and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaShanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, China; MOE Key Laboratory of AI, School of Electronic, Information, and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaShanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, China; MOE Key Laboratory of AI, School of Electronic, Information, and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaShanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, ChinaDepartment of Ophthalmology, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, ChinaDepartment of Ophthalmology, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, ChinaShanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, China; MOE Key Laboratory of AI, School of Electronic, Information, and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaDepartment of Clinical Laboratory, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, ChinaDepartment of Nephrology, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, ChinaDepartment of Pathology, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, ChinaMOE Key Laboratory of AI, School of Electronic, Information, and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaShanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, China; MOE Key Laboratory of AI, School of Electronic, Information, and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaCentre for Public Health, School of Medicine, Dentistry, and Biomedical Sciences, Queen's University Belfast, Belfast, UKCentre for Public Health, School of Medicine, Dentistry, and Biomedical Sciences, Queen's University Belfast, Belfast, UKWestmead Institute for Medical Research, University of Sydney, NSW, AustraliaSchool of Computing, National University of Singapore, SingaporeSchool of Computing, National University of Singapore, SingaporeSchool of Computing, National University of Singapore, SingaporeDepartment of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong Special Administrative Region, ChinaDepartment of Medicine and Therapeutics, Hong Kong Institute of Diabetes and Obesity, Li Ka Shing Institute of Health Sciences, The Chinese University of Hong Kong, Prince of Wales Hospital, Hong Kong Special Administrative Region, ChinaDepartment of Medicine and Therapeutics, Hong Kong Institute of Diabetes and Obesity, Li Ka Shing Institute of Health Sciences, The Chinese University of Hong Kong, Prince of Wales Hospital, Hong Kong Special Administrative Region, ChinaDepartment of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong Special Administrative Region, ChinaSingapore Eye Research Institute, Singapore National Eye Centre, SingaporeSingapore Eye Research Institute, Singapore National Eye Centre, Singapore; Centre for Innovation and Precision Eye Health, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; Ophthalmology and Visual Science Academic Clinical Program, Duke-NUS Medical School, SingaporeSingapore Eye Research Institute, Singapore National Eye Centre, Singapore; Centre for Innovation and Precision Eye Health, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; Ophthalmology and Visual Science Academic Clinical Program, Duke-NUS Medical School, SingaporeSingapore Eye Research Institute, Singapore National Eye Centre, Singapore; Ophthalmology and Visual Science Academic Clinical Program, Duke-NUS Medical School, SingaporeDepartment of Medicine, Faculty of Medicine, Universiti Malaya, Kuala Lumpur, Malaysia; Department of Medicine and Therapeutics, Hong Kong Institute of Diabetes and Obesity, Li Ka Shing Institute of Health Sciences, The Chinese University of Hong Kong, Prince of Wales Hospital, Hong Kong Special Administrative Region, ChinaShanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, ChinaShanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, China; Prof Huating Li, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai 200233, ChinaShanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Institute for Proactive Healthcare, Shanghai Jiao Tong University, Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Centre for Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, China; MOE Key Laboratory of AI, School of Electronic, Information, and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, China; Prof Bin Sheng, Department of Computer Science and Engineering, School of Electronic, Information, and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, ChinaBeijing Visual Science and Translational Eye Research Institute (BERI), Beijing Tsinghua Changgung Hospital Eye Center, School of Clinical Medicine, Tsinghua Medicine, Tsinghua University, Beijing, China; Singapore Eye Research Institute, Singapore National Eye Centre, Singapore; Beijing Key Laboratory of Intelligent Diagnostic Technology and Devices for Major Blinding Eye Diseases, Tsinghua Medicine, Tsinghua University, Beijing, China; Correspondence to: Prof Tien Yin Wong, Beijing Visual Science and Translational Eye Research Institute (BERI), Beijing Tsinghua Changgung Hospital Eye Center, School of Clinical Medicine, Tsinghua Medicine, Tsinghua University, Beijing 100084, ChinaSummary: Background: Improving the accessibility of screening diabetic kidney disease (DKD) and differentiating isolated diabetic nephropathy from non-diabetic kidney disease (NDKD) are two major challenges in the field of diabetes care. We aimed to develop and validate an artificial intelligence (AI) deep learning system to detect DKD and isolated diabetic nephropathy from retinal fundus images. Methods: In this population-based study, we developed a retinal image-based AI-deep learning system, DeepDKD, pretrained using 734 084 retinal fundus images. First, for DKD detection, we used 486 312 retinal images from 121 578 participants in the Shanghai Integrated Diabetes Prevention and Care System for development and internal validation, and ten multi-ethnic datasets from China, Singapore, Malaysia, Australia, and the UK (65 406 participants) for external validation. Second, to differentiate isolated diabetic nephropathy from NDKD, we used 1068 retinal images from 267 participants for development and internal validation, and three multi-ethnic datasets from China, Malaysia, and the UK (244 participants) for external validation. Finally, we conducted two proof-of-concept studies: a prospective real-world study with 3 months' follow-up to evaluate the effectiveness of DeepDKD in screening DKD; and a longitudinal analysis of the effectiveness of DeepDKD in differentiating isolated diabetic nephropathy from NDKD on renal function changes with 4·6 years' follow-up. Findings: For detecting DKD, DeepDKD achieved an area under the receiver operating characteristic curve (AUC) of 0·842 (95% CI 0·838–0·846) on the internal validation dataset and AUCs of 0·791–0·826 across external validation datasets. For differentiating isolated diabetic nephropathy from NDKD, DeepDKD achieved an AUC of 0·906 (0·825–0·966) on the internal validation dataset and AUCs of 0·733–0·844 across external validation datasets. In the prospective study, compared with the metadata model, DeepDKD could detect DKD with higher sensitivity (89·8% vs 66·3%, p<0·0001). In the longitudinal study, participants with isolated diabetic nephropathy and participants with NDKD identified by DeepDKD had a significant difference in renal function outcomes (proportion of estimated glomerular filtration rate decline: 27·45% vs 52·56%, p=0·0010). Interpretation: Among diverse multi-ethnic populations with diabetes, a retinal image-based AI-deep learning system showed its potential for detecting DKD and differentiating isolated diabetic nephropathy from NDKD in clinical practice. Funding: National Key R &amp; D Program of China, National Natural Science Foundation of China, Beijing Natural Science Foundation, Shanghai Municipal Key Clinical Specialty, Shanghai Research Centre for Endocrine and Metabolic Diseases, Innovative research team of high-level local universities in Shanghai, Noncommunicable Chronic Diseases-National Science and Technology Major Project, Clinical Special Program of Shanghai Municipal Health Commission, and the three-year action plan to strengthen the construction of public health system in Shanghai.http://www.sciencedirect.com/science/article/pii/S2589750025000408
spellingShingle Ziyao Meng, MEng
Zhouyu Guan, MD
Shujie Yu, MBBS
Yilan Wu, BSc
Yaoning Zhao, BSc
Jie Shen, PhD
Cynthia Ciwei Lim, MMed
Tingli Chen, MD
Dawei Yang, PhD
An Ran Ran, PhD
Feng He, MSc
Haslina Hamzah, BSc
Sarkaaj Singh, MSc
Anis Syazwani Abd Raof, MSc
Jian Wen Samuel Lee-Boey, MBBS
ProfMBBS Soo-Kun Lim, MBBS
ProfMD Xufang Sun, MBBS
Shuwang Ge, MD
ProfMD Gang Xu, MD
Prof Hua Su, MD
Yang Cheng, MD
Feng Lu, PhD
ProfPhD Xiaofei Liao, PhD
ProfPhD Hai Jin, PhD
Chenxin Deng, MMed
Lei Ruan, MD
ProfMD Cuntai Zhang, MD
Chan Wu, MD
ProfMD Rongping Dai, MD
Yixiao Jin, MSc
Wenxiao Wang, PhD
Tingyao Li, BSc
Ruhan Liu, PhD
Jiajia Li, PhD
Jia Shu, BEng
Yuwei Lu, MBBS
Xiangning Wang, MD
Qiang Wu, MD
Yiming Qin, BEng
Jin Tang, MD
Xiaohua Sheng, MD
Qiong Jiao, MD
ProfPhD Xiaokang Yang, MD
ProfPhD Minyi Guo, MD
Gareth J McKay, PhD
ProfPhD Ruth E Hogg, PhD
Gerald Liew, MBBS
Evelyn Yi Lyn Chee, PhD
ProfPhD Wynne Hsu, PhD
ProfPhD Mong Li Lee, PhD
Simon Szeto, MBChB
ProfMD Andrea O Y Luk, MBChB
ProfMD Juliana C N Chan, MBChB
Carol Y Cheung, PhD
Gavin Siew Wei Tan, FRCSEd
Yih-Chung Tham, PhD
ProfMD PhD Ching-Yu Cheng, PhD
Charumathi Sabanayagam, MD PhD
ProfPhD Lee-Ling Lim, MD PhD
ProfMD PhD Weiping Jia, MD PhD
ProfMD PhD Huating Li, MD PhD
Bin Sheng, Prof
Tien Yin Wong, Prof
Non-invasive biopsy diagnosis of diabetic kidney disease via deep learning applied to retinal images: a population-based study
The Lancet: Digital Health
title Non-invasive biopsy diagnosis of diabetic kidney disease via deep learning applied to retinal images: a population-based study
title_full Non-invasive biopsy diagnosis of diabetic kidney disease via deep learning applied to retinal images: a population-based study
title_fullStr Non-invasive biopsy diagnosis of diabetic kidney disease via deep learning applied to retinal images: a population-based study
title_full_unstemmed Non-invasive biopsy diagnosis of diabetic kidney disease via deep learning applied to retinal images: a population-based study
title_short Non-invasive biopsy diagnosis of diabetic kidney disease via deep learning applied to retinal images: a population-based study
title_sort non invasive biopsy diagnosis of diabetic kidney disease via deep learning applied to retinal images a population based study
url http://www.sciencedirect.com/science/article/pii/S2589750025000408
work_keys_str_mv AT ziyaomengmeng noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT zhouyuguanmd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT shujieyumbbs noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT yilanwubsc noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT yaoningzhaobsc noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT jieshenphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT cynthiaciweilimmmed noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT tinglichenmd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT daweiyangphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT anranranphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT fenghemsc noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT haslinahamzahbsc noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT sarkaajsinghmsc noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT anissyazwaniabdraofmsc noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT jianwensamuelleeboeymbbs noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profmbbssookunlimmbbs noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profmdxufangsunmbbs noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT shuwanggemd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profmdgangxumd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profhuasumd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT yangchengmd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT fengluphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profphdxiaofeiliaophd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profphdhaijinphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT chenxindengmmed noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT leiruanmd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profmdcuntaizhangmd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT chanwumd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profmdrongpingdaimd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT yixiaojinmsc noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT wenxiaowangphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT tingyaolibsc noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT ruhanliuphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT jiajialiphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT jiashubeng noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT yuweilumbbs noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT xiangningwangmd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT qiangwumd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT yimingqinbeng noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT jintangmd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT xiaohuashengmd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT qiongjiaomd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profphdxiaokangyangmd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profphdminyiguomd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT garethjmckayphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profphdruthehoggphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT geraldliewmbbs noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT evelynyilyncheephd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profphdwynnehsuphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profphdmonglileephd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT simonszetombchb noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profmdandreaoylukmbchb noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profmdjulianacnchanmbchb noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT carolycheungphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT gavinsiewweitanfrcsed noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT yihchungthamphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profmdphdchingyuchengphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT charumathisabanayagammdphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profphdleelinglimmdphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profmdphdweipingjiamdphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT profmdphdhuatinglimdphd noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT binshengprof noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy
AT tienyinwongprof noninvasivebiopsydiagnosisofdiabetickidneydiseaseviadeeplearningappliedtoretinalimagesapopulationbasedstudy