An implementation of the paper "Arrhythmia detection using deep convolutional neural network with long duration ECG signals"
Create Arrhythmia Dataset.py generates the task dataset from the original MIT-BIH-Arrhythmia Dataset. This function follows the description in the paper, and extracts arrhythmia 12 classes labeled both by number and by a string
Arrhythmia_Classifier.py contains the CNN according to the details in the paper.
TODO:
- Split the dataset to train and test
- Write classification scores