Detection and Risk Prediction of Diabetes Related Kidney Disease Using Retinal Images: A Systematic Review and Meta-Analysis.
7 affiliations
- 1Centre for Intelligent Healthcare, Coventry University, Coventry, UK.
- 2Institute for Physical Activity and Nutrition, School of Exercise and Nutrition Sciences, Deakin University, Melbourne, Victoria, Australia.
- 3King's Ophthalmology Research Unit (KORU), Section of Ophthalmology, Faculty of Life Sciences and Medicine, King's College London, London, UK.
- 4School of Computing, Engineering & Digital Technologies, Teesside University, Middlesbrough, UK.
- 5Baker Heart and Diabetes Institute, Melbourne, Victoria, Australia.
- 6Diabetes Research Group, Medical School, Swansea University, Swansea, UK.
- 7Melbourne School of Population and Global Health, University of Melbourne, Melbourne, Victoria, Australia.
Abstract
Aim
This study aimed to determine the effectiveness of deep learning models for detecting and predicting diabetes related kidney disease (DRKD) using two-dimensional retinal images.Methods
A systematic review was conducted across eight databases from inception to July 14, 2026. Eligible cross-sectional and longitudinal studies assessed deep learning models for DRKD detection or prediction using retinal images alone, clinical data alone, or hybrid models. Data on study design, model development, validation, and performance metrics were extracted, with study quality appraised using QUADAS-2 and post hoc, PROBAST+AI. Results were synthesized narratively, with random effects meta analysis for hybrid models.Results
Of 367 studies, six met inclusion criteria. Three cross-sectional studies of hybrid models contributed to the meta-analysis for DRKD detection, yielding pooled AUC 0.88, sensitivity 0.80, and specificity 0.81, though heterogeneity was high (I2 93.7%-99.8%). Three further studies were synthesized narratively rather than pooled: two cross-sectional studies used more severe or narrower CKD case definitions (AUC 0.86 and 0.96 respectively), and only one longitudinal study predicted the 5-year risk of DRKD (AUC 0.81, sensitivity 0.70, and specificity 0.77). External validation showed hybrid models consistently outperformed retinal-only (ΔAUC 0.03-0.11), but their advantage over clinical-only models was inconsistent.Conclusion
Deep learning models may be useful for detecting and predicting DRKD from retinal images, with or without clinical data. However, the small number of heterogenous studies restricts generalizability. Future work should prioritize robust model development, and validation using large, diverse populations.
Related
- A systematic review of prognostic multivariable risk prediction models for type 2 diabetes in prediabetic populations
- The effects of improved access to diabetes medications on risk of Alzheimer's disease and related dementias
- Development and validation of a novel metabolic-inflammatory health-related nomogram for predicting in-hospital mortality in older patients with type 2 diabetes mellitus combined with cardiovascular disease
- Diabetes‐Related Foot Disease as the Principal Reason for Hospital Admission: 5 Years of Data From a Major Tertiary Hospital in Aotearoa New Zealand
- Corneal endothelial changes in type 2 diabetes mellitus: a systematic review and meta-analysis with subgroup comparisons by diabetic retinopathy status and disease duration
BibTeX
@article{otabir2026detection,
title = {{Detection and Risk Prediction of Diabetes Related Kidney Disease Using Retinal Images: A Systematic Review and Meta-Analysis.}},
author = {Samuel Ato Gyasi Otabir and Teketo Kassaw Tegegne and Paul Nderitu and Jiangtao Wang and Jonathan E Shaw and David Owens and Rebecca Thomas and Robyn Jennifer Tapp and Ralph Maddison},
journal = {Journal of diabetes},
year = {2026},
month = oct,
doi = {10.1111/1753-0407.70271},
url = {https://europepmc.org/article/MED/42895056},
note = {CC BY},
}
RIS
TY - JOUR
TI - Detection and Risk Prediction of Diabetes Related Kidney Disease Using Retinal Images: A Systematic Review and Meta-Analysis.
AU - Samuel Ato Gyasi Otabir
AU - Teketo Kassaw Tegegne
AU - Paul Nderitu
AU - Jiangtao Wang
AU - Jonathan E Shaw
AU - David Owens
AU - Rebecca Thomas
AU - Robyn Jennifer Tapp
AU - Ralph Maddison
PY - 2026
DA - 2026/10/01
JO - Journal of diabetes
DO - 10.1111/1753-0407.70271
UR - https://europepmc.org/article/MED/42895056
AB - Aim This study aimed to determine the effectiveness of deep learning models for detecting and predicting diabetes related kidney disease (DRKD) using two-dimensional retinal images.Methods A systematic review was conducted across eight databases from inception to July 14, 2026. Eligible cross-sectional and longitudinal studies assessed deep learning models for DRKD detection or prediction using retinal images alone, clinical data alone, or hybrid models. Data on study design, model development, validation, and performance metrics were extracted, with study quality appraised using QUADAS-2 and post hoc, PROBAST+AI. Results were synthesized narratively, with random effects meta analysis for hybrid models.Results Of 367 studies, six met inclusion criteria. Three cross-sectional studies of hybrid models contributed to the meta-analysis for DRKD detection, yielding pooled AUC 0.88, sensitivity 0.80, and specificity 0.81, though heterogeneity was high (I2 93.7%-99.8%). Three further studies were synthesized narratively rather than pooled: two cross-sectional studies used more severe or narrower CKD case definitions (AUC 0.86 and 0.96 respectively), and only one longitudinal study predicted the 5-year risk of DRKD (AUC 0.81, sensitivity 0.70, and specificity 0.77). External validation showed hybrid models consistently outperformed retinal-only (ΔAUC 0.03-0.11), but their advantage over clinical-only models was inconsistent.Conclusion Deep learning models may be useful for detecting and predicting DRKD from retinal images, with or without clinical data. However, the small number of heterogenous studies restricts generalizability. Future work should prioritize robust model development, and validation using large, diverse populations.
SN - 1753-0393
SN - 1753-0407
ER -