Arrhythmias & electrophysiology
Identification of Brugada syndrome based on P-wave features: an artificial intelligence-based approach
Authors' conclusion
An AI machine-learning model is able to identify patients with BrS based only on P-wave characteristics.
Study context
- Population
- Patients with spontaneous or ajmaline-induced brugada syndrome (BrS) and subjects with suspected BrS and negative ajmaline challenge; 5 min 12-lead ECG in sinus rhythm2,228 averaged P-wave observations
- Participants
- 123 patients
- Comparison
- BrS+ vs BrS-
- BrS+n = 79
- BrS-n = 44reference
- Outcome
- Identification of BrS based exclusively on P-wave features by an AI-based model
Methodological detailsArtificial intelligence (AI)-based supervised classifiers, patient-wise 80%/20% train/test split
- AI-based supervised classifiers
- Patient-wise 80%/20% train/test split
Performance
As reported in the abstract
Classification of BrS from P-wave features
| Test or model | Accuracy | Sensitivity | Specificity |
|---|---|---|---|
| AdaBoost | over 81% | 86% | 73% |
AdaBoost
Accuracyover 81%
Sensitivity86%
Specificity73%
Further findings
| Outcome | BrS+ | BrS- | P value |
|---|---|---|---|
| P-wave durationBrS+ vs BrS- | 136 ms | 124 ms | P < 0.001 |
| P-wave terminal force in lead V1BrS+ vs BrS- | 2.5 au | 1.7 au | P < 0.01 |
Original publication
Identification of Brugada syndrome based on P-wave features: an artificial intelligence-based approach. Europace. Published 7 November 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →