Heart failure & cardiomyopathies
Time-Domain Heart Rate Variability Features for Automatic Congestive Heart Failure Prediction
Authors' conclusion
The results show that HRV can accurately predict severe congestive heart failure.
Study context
- Population
- Adults with normal sinus rhythm and adults with congestive HF (NYHA classes 1, 2 and 3); ECG recordings from PhysioNet54 normal sinus rhythm
- Comparison
- Machine learning classifiers (SVM, k-nearest neighbour (KNN), naïve Bayes, decision tree (DT)) on time-domain heart rate variability (HRV) features for HF vs normal sinus rhythm
- Outcome
- Classification of healthy adults vs HF from HRV data (accuracy, precision, recall, specificity, F1 score, AUC)
Methodological detailsfive-fold cross-validation, PhysioNet database recordings
- Five-fold cross-validation
- PhysioNet database recordings
Performance
As reported in the abstract
Classification of HF vs healthy (5 min HRV window)
| Test or model | AUC | Accuracy |
|---|---|---|
| KNN | 0.77 | 77% |
| DT | 0.78 | 77% |
AUC
KNN0.77
DT0.78
Accuracy
KNN77%
DT77%
Further findings
| Outcome | KNN | DT |
|---|---|---|
| Classification of severe congestive HF | ||
| Accuracy | 91% | 91% |
| AUC | 0.88 | 0.92 |
Original publication
Time-Domain Heart Rate Variability Features for Automatic Congestive Heart Failure Prediction. ESC Heart Fail. Published 27 November 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →