Heart failure & cardiomyopathiesCohort study
Artificial intelligence-enabled electrocardiography and echocardiography to track preclinical progression of transthyretin amyloid cardiomyopathy
- Retrospective cohort study
- internal and external cohort
Authors' conclusion
Artificial intelligence-enabled echocardiography and electrocardiography may enable scalable risk stratification of ATTR-CM during its preclinical course.
Study context
- Population
- Individuals referred for nuclear cardiac amyloid testing at Yale-New Haven Health System (YNHHS) (internal) and Houston Methodist Hospitals (HMH) (external)984 YNHHS, 806 HMH
- Exposure
- AI-Echo and AI-ECG screening for preclinical ATTR-CM
Main finding
As reported in the abstract
Longitudinal AI-derived ATTR-CM probabilities, cases vs controls
Artificial intelligence (AI)-derived probabilities (7352 TTEs, 32 205 electrocardiography (ECGs)) diverged as early as 3 years before diagnosis in cases vs controls
ptime(x)group interaction ≤ .004
Observational data: associations, not effects.
Further findings
| Outcome | Double-negative screen | Double-positive screen |
|---|---|---|
| Screen performance 1 to 3 years before nuclear testing (YNHHS n = 433; HMH n = 174) · vs all else | ||
| Sensitivity (YNHHS, HMH) | 90.9% and 85.7% | 60.6% and 42.9% |
| Specificity (YNHHS, HMH) | 40.3% and 41.2% | 85.5% and 88.9% |
| Screened (YNHHS, HMH) | 164 (37.9%) and 66 (37.9%) | 78 (18.0%) and 26 (14.9%) |
- Screen performance 1 to 3 years before nuclear testing (YNHHS n = 433; HMH n = 174): 0.05 threshold
ATTR-CM positive
- YNHHS
- 112 (11.4%)
- HMH
- 174 (21.6%)
Original publication
Artificial intelligence-enabled electrocardiography and echocardiography to track preclinical progression of transthyretin amyloid cardiomyopathy. Eur Heart J. Published 18 July 2025. doi:10.1093/eurheartj/ehaf450
Summarised from the published abstract; every number is checked against it