Heart failure & cardiomyopathies
Impact of Case and Control Selection on Training Artificial Intelligence Screening of Cardiac Amyloidosis
Authors' conclusion
Models performed similarly in population screening, regardless of stringency of cases used during training, showing that institutions without dedicated amyloid clinics can train meaningful models on less curated CA cases.
Study context
- Population
- Patients with electrocardiogram (ECGs) (∼1.3 million ECGs) screened for cardiac amyloidosis; Cedars-Sinai general patient population test cohort
- Participants
- 341,989 patients∼1.3 million ECGs
- Comparison
- ECG-based AI models trained with different case and control definitions for cardiac amyloidosis (CA) screening
- Outcome
- AUC of ECG-based AI models for CA screening
Performance
As reported in the abstract
AUC across models
AUC range, matched held-out test sets
0.660 (95% CI: 0.642-0.736) to 0.898 (95% CI: 0.868-0.924)
AUC range, Cedars-Sinai general population cohort
0.467 (95% CI: 0.443-0.491) to 0.898 (95% CI: 0.870-0.923)
Original publication
Impact of Case and Control Selection on Training Artificial Intelligence Screening of Cardiac Amyloidosis. JACC Adv. Published 12 June 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →