Cardiac arrhythmia refers to any abnormality in beating of the heart. There are many types of cardiac arrhythmia ranging in severity, including premature beats, atrial fibrillation, and ventricular fibrillation. While arrhythmia classification has been well researched, this study focuses on the use of recent techniques in deep learning to classify arrhythmia with minimal possible data pre-processing. Using convolutional neural networks and long short-term memory recurrent neural networks, we were able to produce accuracies of 99.15% and 98.59% respectively.
More details in the report.