Hip Fracture Risk Assessment in Elderly and Diabetic Patients: Combining Autonomous Finite Element Analysis and Machine Learning
Authors' conclusion
Combining AFE data with a ML algorithm provides an unprecedented prediction accuracy for the risk of hip fracture in T2DM and non-T2DM populations.
Study context
- Population
- Patients with abdominal/pelvic CT scans at a tertiary medical center: hip fracture within 2 years after index CT vs no known hip fracture for ≥5 yearsfemurs analysed
- Participants
- 836 patients
- Comparison
- AFE + machine learning (ML) hip fracture prediction in T2D vs non-T2D patients
- T2DMn = 282
- non-T2DMn = 554
- Outcome
- Prediction of hip fracture (sensitivity, specificity, AUC)
Methodological detailsopportunistic, single tertiary medical centre, SVM trained on 80%, cross-validation, verified on 20%, T2DM femurs: 118 intact, 164 fractured; non-T2DM: 314 intact, 240 fractured
- Opportunistic
- Single tertiary medical centre
- SVM trained on 80%, cross-validation, verified on 20%
- T2DM femurs: 118 intact, 164 fractured; non-T2DM: 314 intact, 240 fractured
Performance
As reported in the abstract
Hip fracture prediction
| Test or model | Cross-validation AUC | Sensitivity | Specificity |
|---|---|---|---|
| T2DM | 0.92 | 92% | 88% |
| non-T2DM | 0.84 | 83% | 84% |
Cross-validation AUC
T2DM0.92
non-T2DM0.84
Sensitivity
T2DM92%
non-T2DM83%
Specificity
T2DM88%
non-T2DM84%
Further findings
Feasibility of AFE
- CT scans appropriate for AFE
- 45%
- AFE success rate
- 91%
Original publication
Hip Fracture Risk Assessment in Elderly and Diabetic Patients: Combining Autonomous Finite Element Analysis and Machine Learning. J Bone Miner Res. Published 27 March 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →