This repository aims to collect open source implementations of ML methods in cardiology to make them available for benchmarking. Its structure mirrors the two fundamental classification tasks relevant for the field: Beat classification and Phenotype classification.
Available implementations:
- [keras] Paper: Screening for cardiac contractile dysfunction using an artificial intelligence–enabled electrocardiogram (Attia et al.)
Open implementations (non-exhaustive):
- Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network (Hannun et al.): Github repository
- Real-Time Patient-Specific ECG Classification by 1-D Convolutional Neural Networks (Kiranyaz et al.)
If you would like to contribute an implementation, feel free to open a pull request.