Heart failure & cardiomyopathies
Echocardiographic Tissue Characterization Using Radiomics in Patients With Transthyretin-Related Cardiac Amyloidosis
Authors' conclusion
This study showcases ultrasonomics potential to differentiate ATTR-CA and control patients by capturing disease-specific textural features independent of ROI dimensions.
Study context
- Population
- ATTR-CA and control patients with echocardiographic images in parasternal long-axis and apical 4-chamber viewsDevelopment, 3 centers: ATTR-CA 229 · Development, 3 centers: controls 224 · External validation, fourth center: ATTR-CA 32 · External validation, fourth center: controls 32
- Comparison
- Logistic regression models of echocardiographic radiomic features of the interventricular septum region of interest (ROI) to differentiate ATTR-CA and control patients
Methodological details4 Italian centers, external validation
- 4 Italian centers
- External validation
Further findings
Diagnostic performance, entire ROI with ROI-dependent and ROI-independent features
- Cross-validated accuracies
- 93%-95%
- AUC
- 0.97-0.99
Diagnostic performance, fixed-size 0.5 × 0.5 cm ROI
- Accuracy
- 85%
- AUC
- 0.91
- Highlighting previous models' dependence on ROI size
Fifth model: 73 ROI-independent features on the entire ROI
- Accuracy
- 92%
- AUC
- 0.97
- P
- P < 0.001
- External validation accuracy
- 87%
- External validation AUC
- 0.95
- Significantly better accuracy and AUC
Model after removing the least informative features
- Accuracy
- 90%
- Precision
- 0.95
Original publication
Echocardiographic Tissue Characterization Using Radiomics in Patients With Transthyretin-Related Cardiac Amyloidosis. JACC Adv. Published 3 May 2025. doi:10.
Summarised from the published abstract; every number is checked against it