Heart failure & cardiomyopathiesCohort study
AI-ECG for Detecting Left Ventricular Systolic Dysfunction in Chagas Disease
- Brazilian prospective Chagas disease (ChD) cohort
- previously developed AI-ECG left ventricular systolic dysfunction (LVSD) model fine-tuned in an external ChD sample
Authors' conclusion
A fine-tuned AI-ECG model showed high accuracy for LVSD detection and was independently associated with long-term mortality and incident LVSD. Despite slightly lower performance than NT-proBNP, it may serve as a substitute in settings where NT-proBNP is unavailable. (Longitudinal Study of Patients With Chronic Chagas Cardiomyopathy in Brazil [SaMi_Trop Project] [SaMi-Trop]; NCT02646943).
Study context
- Population
- Patients with Chagas disease (ChD) in the SaMi-Trop cohortparticipants in the LVSD detection analysis, 1,547 patients with Chagas cardiomyopathy in the mortality analysis · LVSD detection 1,304 · Chagas cardiomyopathy, mortality prediction 1,547
- Participants
- 1,304 patients
- Exposure
- Diagnostic performance for LVSD, confirmed by echocardiography, was compared with N-terminal pro-B-type natriuretic peptide (NT-proBNP)
Main finding
As reported in the abstract
AI-ECG
0.89
AUC · 95% CI 0.85–0.93
NT-proBNP
0.90
AUC · 95% CI 0.87–0.93
Observational data: associations, not effects.
Further findings
C-index
| Outcome | AI-ECG | NT-proBNP |
|---|---|---|
| All-cause mortality · 9 years | 0.78 (95% CI 0.74–0.82) | 0.77 (95% CI 0.74–0.79) |
- All-cause mortality: Among 1,547 patients with Chagas cardiomyopathy
Incident LVSD · 7.3 years
- Incident LVSD occurred
- 8.4%
- AI-ECG predicted incident LVSD with performance similar to NT-proBNP
Replacement of NT-proBNP in an established ChD mortality risk score
- AI-ECG could replace NT-proBNP in an established ChD mortality risk score with minor loss of accuracy
Original publication
Summarised from the published abstract; every number is checked against it