A Machine Learning Model for Prediction of Amputation in Diabetics
Authors' conclusion
Utilizing machine learning methods, we have developed a clinical algorithm that predicts the risk of major lower extremity amputation for inpatients with diabetes with 77.8% accuracy.
Study context
- Population
- Hospitalized patients with diabetic foot ulcer (DFU), National Inpatient Sample 2008–2014inpatients with DFU
- Participants
- 326,853 patients
- Outcome
- Prediction of major amputation
Methodological detailsretrospective, National Inpatient Sample database 2008–2014, decision tree (CTREE) and random forest models
- Retrospective
- National Inpatient Sample database 2008–2014
- Decision tree (CTREE) and random forest models
Performance
As reported in the abstract
Model performance for predicting major amputation
Accuracy, training data (76.1% sensitivity, 79.3% specificity)
77.7%
Accuracy, testing data (76.2% sensitivity, 79.4% specificity)
77.8%
AUC, boosting and random forest validation
0.84
95% CI 0.83–0.85
Further findings
Major amputation rate
- Underwent major amputation
- 5.9%
Top five predictors of major amputation (all P < .001)
- Gangrene, OR
- 11.8 (95% CI 11.5-12.2)
- Peripheral vascular disease, OR
- 2.9 (95% CI 2.8-3.0)
- Weight loss, OR
- 2.6 (95% CI 2.5-2.8)
- Systemic infection, OR
- 2.5 (95% CI 2.4-2.53)
- Osteomyelitis, OR
- 1.7 (95% CI 1.6-1.73)
Original publication
A Machine Learning Model for Prediction of Amputation in Diabetics. J Diabetes Sci Technol. Published 8 December 2022. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →