Heart failure & cardiomyopathies
Deep learning model for identifying acute heart failure patients using electrocardiography in the emergency room
Authors' conclusion
The developed machine learning model significantly enhanced AHF detection in the ER using conventional 12-lead ECGs combined with clinical data. These findings suggest that ECGs, a common tool in the ER, can effectively help screen for AHF.
Study context
- Population
- Patients who visited emergency room (ERs), with available left ventricular ejection fraction and N-terminal prohormone of brain natriuretic peptide level dataincluded, of 19,285 patients whose electrocardiogram (ECG) data were analysed
- Participants
- 9,119 patients
- Index test or model
- Machine learning models using morphological and clinical parameters extracted from ECG data for acute heart failure (AHF) identification in the ER
- Target condition
- Acute heart failure (AHF)
Methodological detailsthree hospitals
- Three hospitals
Performance
As reported in the abstract
| Test or model | Area under the receiver operating characteristic | Area under the precision-recall curve |
|---|---|---|
| CatBoost, internal validation | 0.89 ± 0.01 | 0.89 ± 0.01 |
| CatBoost, external validation | 0.90 | 0.89 |
Area under the receiver operating characteristic
CatBoost, internal validation0.89 ± 0.01
CatBoost, external validation0.90
Area under the precision-recall curve
CatBoost, internal validation0.89 ± 0.01
CatBoost, external validation0.89
Further findings
- The model demonstrated high accuracy, precision, recall, and f1 score
Original publication
Deep learning model for identifying acute heart failure patients using electrocardiography in the emergency room. Eur Heart J Acute Cardiovasc Care. Published 9 January 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →