Heart failure & cardiomyopathiesDiagnostic or prognostic study
Multimodal Artificial Intelligence for Cardiac Amyloidosis Diagnosis: Integrating Echocardiography With Clinical and Laboratory Data for Improved Detection
- developed and tested a combined artificial intelligence (AI) echo-clinical model
- global multiethnic registry
- internal validation
Authors' conclusion
A multiparametric AI model integrating basic clinical, laboratory, and TTE data with the deep learning Us2.Ca improved performance for CA detection over Us2.Ca alone. This approach represents a step toward scalable, AI-guided precision diagnostics for CA in diverse populations.
Study context
- Population
- Patients with cardiac amyloidosis and controls from the Amyloidosis Imaging International Consortium727 patients with cardiac amyloidosis (CA) and 316 controls · patients with CA 727 · controls 316 · controls: suspected transthyretin-CA with negative diagnostic evaluation 202 · controls: biopsy-proven extracardiac light chain amyloidosis without cardiac involvement 114
- Comparison
- AI echo-clinical model (AI-ECM) vs Us2.Ca
Performance
As reported in the abstract
Diagnostic performance for CA detection
| Test or model | area under the curve | accuracy | sensitivity | specificity |
|---|---|---|---|---|
| AI-ECM | 0.94 | 90% | 93% | 85% |
| Us2.Ca | 0.89 | 80% | 76% | 91% |
- AI-ECM vs Us2.Ca: P = 0.006 — superior performance of AI-ECM
Further findings
| Outcome | AI-ECM | Us2.Ca |
|---|---|---|
| Indeterminate classification | Not reported in the abstract | indeterminate cases 9% |
Feature importance
- Us2.Ca prediction scores, relative wall thickness, sex, and estimated glomerular filtration rate contributed most to performance
Indeterminate classification
- AI-ECM allowed classification of all cases
Subtypes and control groups
- AI-ECM improved sensitivity for light chain-CA detection and maintained high accuracy across subtypes and control groups
Original publication
Multimodal Artificial Intelligence for Cardiac Amyloidosis Diagnosis: Integrating Echocardiography With Clinical and Laboratory Data for Improved Detection. Circ Cardiovasc Imaging. Published 27 May 2026. doi:10.1161/circimaging.126.019610
Summarised from the published abstract; every number is checked against it