Skip to content
DraconiticResearch
Search Sources API
226,742 studies31 sourcesUpdated 42 min ago
Peer-reviewedCC BY-NC 4.0BMJ Health & Care Informatics

Using AI-generated ultra-widefield-like fundus images to assist colour fundus photography in screening for chronic kidney disease: a multicentre cross-sectional study

3 more authors
10 affiliations
  1. 1Department of Ophthalmology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
  2. 2School of Biomedical Engineering, Tsinghua Medicine, Tsinghua University, Beijing, China
  3. 3Department of Ophthalmology, Jiading Hospital of Traditional Chinese Medicine, Shanghai, China
  4. 4Department of Ophthalmology, Nanxiang Hospital of Jiading District, Shanghai, China
  5. 5Department of Ophthalmology, Loujiang New City Hospital, Taicang City (Taicang Branch of Ruijin Hospital), Taicang, Jiangsu, China
  6. 6Department of Hypertension, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
  7. 7The Innovation Center, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
  8. 8School of Biomedical Engineering and the Institute of Medical Robotics, Shanghai Jiao Tong University, Shanghai, China
  9. 9Department of Surgery, Shanghai Key Laboratory of Gastric Neoplasms, Shanghai Institute of Digestive Surgery, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
  10. 10Department of Geriatrics, Medical Center on Aging of Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China

Abstract

Objective To develop a generative artificial intelligence (AI) model that derives ultra-widefield-like (UWF-like) fundus images from conventional colour fundus photographs (CFP) and to assess whether these images improve chronic kidney disease (CKD) screening.Methods and analysis This multicentre cross-sectional diagnostic study included 3776 fundus images from 2056 patients at four centres in China and Japan between 2023 and 2024. The generative model was developed and validated using 1865 images from 1049 patients. CKD screening models were developed using 1750 UWF images and their corresponding UWF-derived CFP-like central crops from 909 patients at Ruijin Hospital with patient-level fivefold cross-validation and externally validated using 161 native CFP images from 98 patients at Jiading Hospital of Traditional Chinese Medicine. Performance was assessed by area under the receiver operating characteristic curve (AUC) and specificity at fixed sensitivity of 0.80.Results In internal validation, AUCs were 0.911 for real UWF images, 0.773 for AI-generated UWF-like images and 0.689 for UWF-derived CFP-like central crops. At fixed sensitivity of 0.80, specificities were 0.879, 0.573 and 0.459, respectively. In external validation, AI-generated UWF-like images outperformed native CFP images, with an AUC of 0.707 versus 0.609 and specificity of 0.458 versus 0.344. Retinal vascular features from AI-generated UWF-like images remained associated with renal function indicators.Conclusion AI-generated UWF-like images improved CKD screening performance compared with CFP alone. This approach may extend retinal imaging-based CKD screening where UWF imaging is unavailable, but prospective real-world validation is needed before clinical implementation.

Related

  1. Comparative detection of diabetic retinopathy microvascular lesions using single-capture ultrawide-field swept-source OCT angiography versus widefield colour fundus photography.
    Yusra Asad, Pavan Paniharam, Ser Koon Goh +17
    BMC ophthalmologyPeer-reviewedCC BY-NC-NDFull textGeneral medicineEurope PMC
  2. Patients’ perspectives on smartphone-based clinical photography in resource-constrained settings: a cross-sectional study
    Urmila Gurung, Ushant Acharya, Khagendra Bhandari +1
    Annals of medicine and surgery (2012)Peer-reviewedCC BYFull textMedical imaging & radiologyEurope PMCEurope PMCPMC
  3. Assessing coping, loneliness, resilience and partner burden in patients with skin conditions: protocol for a multicentre cross-sectional study across 23 dermatology centres in 15 European countries
    Christina Schut, Francesca Sampogna, Jörg Kupfer +3
    BMJ OpenPeer-reviewedCC BY-NC 4.0Full textGeneral medicineCrossrefEurope PMC
  4. Barriers and facilitators to Medical Emergency Team activation among healthcare professionals: protocol for a multicentre cross-sectional survey in four Swiss hospitals
    N. S. Goetz, A. Wolfensberger, R. Ludwig +7
    medRxivPreprintAll rights reservedFull textGeneral medicinemedRxiv
  5. Experiences during medical internship in Bangladesh: a multicentre cross-sectional study.
    Mohammad Jahid Hasan, Sanjoy Kumar Ray, Sadia Islam +9
    BMJ openPeer-reviewedCC BY-NCFull textEpidemiology & public healthEurope PMCCrossref
BibTeX
@article{wu2026using,
  title = {{Using AI-generated ultra-widefield-like fundus images to assist colour fundus photography in screening for chronic kidney disease: a multicentre cross-sectional study}},
  author = {Renkai Wu and Huiping Yao and Shanshan Wang and Yanwei Chen and Linbi Yu and Jingjing Li and Yonghua Kang and Xin Chen and Zhixin Jiang and Yuanlin Mo and Ruiming Fang and Hongen Liao and Guochen Ning and Fang Chen and Qing Chang},
  journal = {BMJ Health \& Care Informatics},
  year = {2026},
  month = nov,
  doi = {10.1136/bmjhci-2026-102263},
  url = {https://doi.org/10.1136/bmjhci-2026-102263},
  note = {CC BY-NC 4.0},
}
RIS
TY  - JOUR
TI  - Using AI-generated ultra-widefield-like fundus images to assist colour fundus photography in screening for chronic kidney disease: a multicentre cross-sectional study
AU  - Renkai Wu
AU  - Huiping Yao
AU  - Shanshan Wang
AU  - Yanwei Chen
AU  - Linbi Yu
AU  - Jingjing Li
AU  - Yonghua Kang
AU  - Xin Chen
AU  - Zhixin Jiang
AU  - Yuanlin Mo
AU  - Ruiming Fang
AU  - Hongen Liao
AU  - Guochen Ning
AU  - Fang Chen
AU  - Qing Chang
PY  - 2026
DA  - 2026/11/11
JO  - BMJ Health & Care Informatics
DO  - 10.1136/bmjhci-2026-102263
UR  - https://doi.org/10.1136/bmjhci-2026-102263
AB  - Objective To develop a generative artificial intelligence (AI) model that derives ultra-widefield-like (UWF-like) fundus images from conventional colour fundus photographs (CFP) and to assess whether these images improve chronic kidney disease (CKD) screening.Methods and analysis This multicentre cross-sectional diagnostic study included 3776 fundus images from 2056 patients at four centres in China and Japan between 2023 and 2024. The generative model was developed and validated using 1865 images from 1049 patients. CKD screening models were developed using 1750 UWF images and their corresponding UWF-derived CFP-like central crops from 909 patients at Ruijin Hospital with patient-level fivefold cross-validation and externally validated using 161 native CFP images from 98 patients at Jiading Hospital of Traditional Chinese Medicine. Performance was assessed by area under the receiver operating characteristic curve (AUC) and specificity at fixed sensitivity of 0.80.Results In internal validation, AUCs were 0.911 for real UWF images, 0.773 for AI-generated UWF-like images and 0.689 for UWF-derived CFP-like central crops. At fixed sensitivity of 0.80, specificities were 0.879, 0.573 and 0.459, respectively. In external validation, AI-generated UWF-like images outperformed native CFP images, with an AUC of 0.707 versus 0.609 and specificity of 0.458 versus 0.344. Retinal vascular features from AI-generated UWF-like images remained associated with renal function indicators.Conclusion AI-generated UWF-like images improved CKD screening performance compared with CFP alone. This approach may extend retinal imaging-based CKD screening where UWF imaging is unavailable, but prospective real-world validation is needed before clinical implementation.
SN  - 2632-1009
ER  -