Diagnosis of coronary artery disease in patients with type 2 diabetes mellitus based on computed tomography and pericoronary adipose tissue radiomics: a retrospective cross-sectional study
Authors' conclusion
A comprehensive diagnostic model combining patient clinical risk factors with CT-based imaging parameters has superior efficacy in diagnosing the occurrence of CAD in patients with T2DM.
Study context
- Population
- Patients with T2D but no coronary artery disease (CAD) history; 146 diagnosed with CAD at this visit, 83 not
- Participants
- 229 patients
- Comparison
- Four diagnostic models: clinical factors (Model 1), + computed tomography (CT) imaging indices (Model 2), + Radscore (Model 3), all together (Model 4)
- Outcome
- Identification of CAD (AUC)
Methodological detailsretrospective, cross-sectional, radiomics, training and test sets (7:3)
- Retrospective
- Cross-sectional
- Radiomics
- Training and test sets (7:3)
Performance
As reported in the abstract
AUC for CAD, test set
| Test or model | AUC |
|---|---|
| Model 1 | 0.691 |
| Model 2 | 0.930 |
| Model 3 | 0.693 |
| Model 4 | 0.929 |
AUC
Model 10.691
Model 20.930
Model 30.693
Model 40.929
- Model 2 vs Model 1: 95% CI 0.129–0.350; P < 0.001 — higher diagnostic efficacy with Model 2; DeLong test
- Model 4 vs Model 2: P = 0.776 — Model 4 did not improve identification
- Model 3 vs Model 1: P = 0.382 — AUC did not significantly differ
- Models 2 and 4 had higher recognition effectiveness than the other two models
- each p < 0.001
Original publication
Diagnosis of coronary artery disease in patients with type 2 diabetes mellitus based on computed tomography and pericoronary adipose tissue radiomics: a retrospective cross-sectional study. Cardiovasc Diabetol. Published 23 January 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →