Machine Learning Models for Prediction of Diabetic Microvascular Complications
Authors' conclusion
There is growing global interest in using ML for predicting diabetic microvascular complications. Research on DKD is the most advanced in terms of publication volume and overall prediction performance. Both DR and DN require more research. External validation and adherence to recommended guidelines are crucial.
Study context
- Population
- Studies of ML/AI models predicting diabetic retinopathy (DR), diabetic kidney disease (DKD) and diabetic neuropathy (DN), PubMed 1990–202374 studies
- Outcome
- Discrimination (c-statistic) of ML prediction models, all microvascular complications
Pooled result
As reported in the abstract
Internal validation, mean (SD) c-statistic
0.79
0.09
External validation, mean (SD) c-statistic
0.72
0.12
Internally validated ML models
256
Externally validated models
124
Further pooled results
Discrimination of DKD models (c-statistic)
- Internal validation
- 0.81 (0.09)
- External validation
- 0.74 (0.13)
- DKD models had the highest discrimination
Determinants of performance and research trends
- Rise in ML use since 2010, mainly DKD research, countries
- 27
- Prediction horizon, outcome definitions, number and type of predictors, and ML technique significantly influenced model performance
- Few studies externally validated prediction of DN
Original publication
Machine Learning Models for Prediction of Diabetic Microvascular Complications. J Diabetes Sci Technol. Published 8 January 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →