DXA based radiomics scoring: a novel approach for fragility fracture risk assessment in elderly type 2 diabetes patients
Authors' conclusion
This DXA-based Radscore presented in this study offers a robust approach to stratify fracture risk in elderly patients with T2DM.
Study context
- Population
- Elderly patients with type 2 diabetes mellitus
- Comparison
- Derivation set vs Validation set
- Derivation setn = 294
- Outcome
- Prediction of fracture risk (vertebral, hip and major osteoporotic fractures) by a DXA lumbar spine and hip radiomic model
Methodological detailstraining cohort and external validation cohort (ABCD cohort), Cox proportional hazards regression, radiomics
- Training cohort and external validation cohort (ABCD cohort)
- Cox proportional hazards regression
- Radiomics
Performance
As reported in the abstract
Discrimination of the DXA radiomics model (AUC)
| Test or model | AUC, radiographically confirmed vertebral fracture | AUC, incident clinical vertebral fracture | AUC, incident hip fracture | AUC, incident MOF |
|---|---|---|---|---|
| Derivation set | 0.830 (95% CI: 0.783-0.878) | — not stated in the abstract | — not stated in the abstract | — not stated in the abstract |
| Validation set | — not stated in the abstract | 0.774 (95% CI: 0.686-0.862) | 0.759 (95% CI: 0.645-0.873) | 0.739 (95% CI: 0.676-0.803) |
AUC, radiographically confirmed vertebral fracture
Derivation set0.830 (95% CI: 0.783-0.878)
Validation set— not stated in the abstract
AUC, incident clinical vertebral fracture
Derivation set— not stated in the abstract
Validation set0.774 (95% CI: 0.686-0.862)
AUC, incident hip fracture
Derivation set— not stated in the abstract
Validation set0.759 (95% CI: 0.645-0.873)
AUC, incident MOF
Derivation set— not stated in the abstract
Validation set0.739 (95% CI: 0.676-0.803)
— not stated in the abstract
Further findings
| Outcome | Derivation set | Validation set |
|---|---|---|
| Fracture occurrence · median 5-year follow-up (validation set) | 35.7% | |
| Radiographically confirmed vertebral fractures | 105 (35.7%) | — not stated in the abstract |
| Incident clinical vertebral fracture | — not stated in the abstract | 3.2% |
| Incident hip fracture | — not stated in the abstract | 2.4% |
| Incident MOFs | — not stated in the abstract | 7.0% |
— not stated in the abstract
Radiomics score (Radscore) relationships and risk stratification
- Weak correlations with traditional fracture assessment tools but independently with the diabetes-specific phenotype
- Stratified T2DM patients into risk groups with significantly differing fracture incidences
Original publication
DXA based radiomics scoring: a novel approach for fragility fracture risk assessment in elderly type 2 diabetes patients. Osteoporos Int. Published 30 July 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →