This project aims to classify ECG signals into four categories, normal rhythm, atrial fibrillation rhythm, other rhythm, and noisy recording. We used The PhysioNet/Computing in Cardiology Challenge 2017 as a primary dataset for this task, but since it does not contain enough samples, we used the MIT-BIH Arrhythmia Database as well.
- Data Cleaning
- Data Balancing
- Data Augmentation
- Python
- Tensorflow/keras
- Pandas
- Numpy
- Matplotlib
- Sklearn
P.s. In June 2020, this project was completed; recent commits have been devoted to organizing it.