机构:[1]Chinese Peoples Liberat Army Gen Hosp, Dept Nephrol, Med Ctr 1, Beijing, Peoples R China[2]Chinese Peoples Liberat Army, Nephrol Inst, Beijing, Peoples R China[3]State Key Lab Kidney Dis, Beijing, Peoples R China[4]Natl Clin Res Ctr Kidney Dis, Beijing, Peoples R China[5]Chinese Peoples Liberat Army Gen Hosp, Beijing, Peoples R China[6]Univ Cambridge, Dept Clin Neurosci, Cambridge, England[7]Beihang Univ, Sch Elect & Informat Engn, Beijing, Peoples R China[8]Chinese Peoples Liberat Army Gen Hosp, Senior Dept Ophthalmol, Med Ctr 3, Beijing, Peoples R China[9]Capital Med Univ, Beijing Tongren Hosp, Beijing Tongren Eye Ctr, Beijing, Peoples R China首都医科大学附属北京同仁医院首都医科大学附属同仁医院[10]Guangdong Pharmaceut Univ, Sch Clin Med, Guangzhou, Peoples R China[11]Chinese Peoples Liberat Army Gen Hosp, Dept Ophthalmol, Med Ctr 1, Beijing, Peoples R China[12]Chinese Peoples Liberat Army Gen Hosp, Dept Pediat, Seventh Med Ctr, Beijing, Peoples R China[13]Capital Med Univ, Beijing Tongren Hosp, Beijing Diabet Inst, Beijing Key Lab Diabet Res & Care, Beijing, Peoples R China研究所糖尿病研究所首都医科大学附属北京同仁医院首都医科大学附属同仁医院[14]Army Med Univ, Daping Hosp, Dept Nephrol, Chongqing, Peoples R China[15]Dalian Med Univ, Ctr Transformat Med Kidney Dis, Key Lab Kidney Dis, Affiliated Hosp 1, Dalian, Peoples R China大连医科大学附属第一医院[16]Chinese Peoples Liberat Army Gen Hosp, Dept Ophthalmol, Med Ctr 3, Beijing, Peoples R China
Differentiating between diabetic nephropathy (DN) and non-diabetic renal disease (NDRD) without a kidney biopsy remains a major challenge, often leading to missed opportunities for targeted treatments that could greatly improve NDRD outcomes. To reform the traditional biopsy-all diagnostic paradigm and avoid unnecessary biopsy, we developed a transformer-based deep learning (DL) system for detecting DN and NDRD upon non-invasive multi-modal data of fundus images and clinical characteristics. Our Trans-MUF achieved an AUC of 0.980 (95% CI: 0.979 to 0.980) over the internal retrospective set and also had superior generalizability over a prospective dataset (AUC: 0.989, 95% CI: 0.987 to 0.990) and a multicenter, cross-machine and multi-operator dataset (AUC: 0.932, 95% CI: 0.931 to 0.939). Moreover, the nephrologists' diagnosis accuracy can be improved by 21%, through visualization assistance of the DL system. This paper lays a foundation for automatically differentiating DN and NDRD without biopsy. (Registry name: Correlation Study Between Clinical Phenotype and Pathology of Type 2 Diabetic Nephropathy. ID: NCT03865914. Date: 2017-11-30).
基金:
Capital’s Funds for Health Improvement
and Research (CFH 2024-1-5021), Science Technology Project of
Beijing, China (No. Z221100007422121), Natural Science Foundation
of China (No.3214100532000530,32371035), NSFC under Grants
62250001, 62231002, and by Beijing Natural Science Foundation under Grant L223021, L232122,7244305, L222133.
语种:
外文
被引次数:
WOS:
PubmedID:
中科院(CAS)分区:
出版当年[2025]版:
大类|1 区医学
小类|1 区卫生保健与服务1 区医学:信息
最新[2025]版:
大类|1 区医学
小类|1 区卫生保健与服务1 区医学:信息
JCR分区:
出版当年[2023]版:
Q1HEALTH CARE SCIENCES & SERVICESQ1MEDICAL INFORMATICS
最新[2024]版:
Q1HEALTH CARE SCIENCES & SERVICESQ1MEDICAL INFORMATICS
第一作者机构:[1]Chinese Peoples Liberat Army Gen Hosp, Dept Nephrol, Med Ctr 1, Beijing, Peoples R China[2]Chinese Peoples Liberat Army, Nephrol Inst, Beijing, Peoples R China[3]State Key Lab Kidney Dis, Beijing, Peoples R China[4]Natl Clin Res Ctr Kidney Dis, Beijing, Peoples R China[5]Chinese Peoples Liberat Army Gen Hosp, Beijing, Peoples R China
共同第一作者:
通讯作者:
通讯机构:[1]Chinese Peoples Liberat Army Gen Hosp, Dept Nephrol, Med Ctr 1, Beijing, Peoples R China[2]Chinese Peoples Liberat Army, Nephrol Inst, Beijing, Peoples R China[3]State Key Lab Kidney Dis, Beijing, Peoples R China[4]Natl Clin Res Ctr Kidney Dis, Beijing, Peoples R China[5]Chinese Peoples Liberat Army Gen Hosp, Beijing, Peoples R China
推荐引用方式(GB/T 7714):
Dong Zheyi,Wang Xiaofei,Pan Sai,et al.A multimodal transformer system for noninvasive diabetic nephropathy diagnosis via retinal imaging[J].NPJ DIGITAL MEDICINE.2025,8(1):doi:10.1038/s41746-024-01393-1.
APA:
Dong, Zheyi,Wang, Xiaofei,Pan, Sai,Weng, Taohan,Chen, Xiaoniao...&Chen, Xiangmei.(2025).A multimodal transformer system for noninvasive diabetic nephropathy diagnosis via retinal imaging.NPJ DIGITAL MEDICINE,8,(1)
MLA:
Dong, Zheyi,et al."A multimodal transformer system for noninvasive diabetic nephropathy diagnosis via retinal imaging".NPJ DIGITAL MEDICINE 8..1(2025)