Heart failure & cardiomyopathies
Pragmatic Screening for Heart Failure in the General Population Using an Electrocardiogram-Based Neural Network
Authors' conclusion
Using a CNN build on digital ECGs only and requiring no additional input, we derived a screening tool for heart failure in the general population.
Study context
- Population
- Randomly selected inhabitants of Hamburg aged 45–75 years318 (6.0%) with HF
- Participants
- 5,299 participants
- Comparison
- ECG-based convolutional neural network (CNN) vs logistic regression and random forest models using HF risk variables
- Outcome
- Discrimination of participants with vs without HF (diagnosed per current guidelines)
Methodological detailsprospective, observational, population-based (Hamburg-City-Health-Study)
- Prospective
- Observational
- Population-based (Hamburg-City-Health-Study)
Performance
As reported in the abstract
HF detection
| Test or model | AUC | Sensitivity | Specificity |
|---|---|---|---|
| CNN | 0.75 | 0.67 | 0.69 |
| Logistic regression | 0.77 | 0.63 | 0.76 |
| Random forest | 0.79 | 0.67 | 0.72 |
AUC
CNN0.75
Logistic regression0.77
Random forest0.79
Sensitivity
CNN0.67
Logistic regression0.63
Random forest0.67
Specificity
CNN0.69
Logistic regression0.76
Random forest0.72
Original publication
Pragmatic Screening for Heart Failure in the General Population Using an Electrocardiogram-Based Neural Network. ESC Heart Fail. Published 8 December 2022. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →