Accuracy of deep learning for risk prediction and screening of diabetic foot Ulcers: A systematic review and meta-analysis
Authors' conclusion
This research discloses that image-based DL for detecting DFUs might be a promising approach. Future studies should further ascertain the generalizability of models derived from public databases to individual-level data across diverse geographic regions, thereby providing evidence for the evaluation and development of AI-assisted tools.
Study context
- Population
- Original studies using image-based deep learning (DL) to detect diabetic foot ulcers (DFUs); PubMed, Cochrane, Embase and Web of Science searched up to October 202555 studies
- Outcome
- Pooled sensitivity and specificity of DL for detecting DFUs
Methodological detailsbivariate mixed-effects model, QUADAS-AI risk of bias
- Bivariate mixed-effects model
- QUADAS-AI risk of bias
Pooled result
As reported in the abstract
Sensitivity, overall validation sets
0.96
95% CI 0.94–0.98
Specificity, overall validation sets
0.97
95% CI 0.94–0.98
Further pooled results
- Sensitivity, multicenter datasets
- 0.87 (95% CI 0.68-0.96)
- Specificity, multicenter datasets
- 0.92 (95% CI 0.82-0.97)
Original publication
Accuracy of deep learning for risk prediction and screening of diabetic foot Ulcers: A systematic review and meta-analysis. Diabetes Res Clin Pract. Published 13 April 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →