Ecg-clip
Intervention1 paper4 findings
- In silico1
Outcomes
4Atrial fibrillation prediction
upin silico1Very low certainty
1 study
Atrial fibrillation prediction
upin silico1Very low certainty
Ecg-clip improves atrial fibrillation prediction in silico (AUC 0.777 (95% CI 0.773-0.781) vs 0.765 [0.761-0.769]).
In silicowith only ten positive labels
Development and external validation of a contrastive learning foundation model for ECG-based prediction of cardiovascular diseases and outcomes · The Lancet. Digital health · 1 Sep 2026
Hypertrophic cardiomyopathy detection
upin silico1Very low certainty
1 study
Hypertrophic cardiomyopathy detection
upin silico1Very low certainty
Ecg-clip improves hypertrophic cardiomyopathy detection in silico (AUC 0.772 (95% CI 0.751-0.793) vs 0.754 [0.733-0.774]).
In silicowith only ten positive labels
Development and external validation of a contrastive learning foundation model for ECG-based prediction of cardiovascular diseases and outcomes · The Lancet. Digital health · 1 Sep 2026
Cardiac amyloidosis detection
upin silico1Very low certainty
1 study
Cardiac amyloidosis detection
upin silico1Very low certainty
Ecg-clip improves cardiac amyloidosis detection in silico (AUC 0.790 (95% CI 0.767-0.812) vs 0.777 [0.754-0.799]).
In silicowith only ten positive labels
Development and external validation of a contrastive learning foundation model for ECG-based prediction of cardiovascular diseases and outcomes · The Lancet. Digital health · 1 Sep 2026
Acute myocardial infarction detection
upin silico1Very low certainty
1 study
Acute myocardial infarction detection
upin silico1Very low certainty
Ecg-clip improves acute myocardial infarction detection in silico (AUC 0.910 (95% CI 0.903-0.916) vs 0.884 [0.876-0.892]).
In silicowith only ten positive labels
Development and external validation of a contrastive learning foundation model for ECG-based prediction of cardiovascular diseases and outcomes · The Lancet. Digital health · 1 Sep 2026