RythmNet: A Novel Convolutional Neural Network for Multiclass ECG Classification

An effective non-invasive diagnostic tool for identifying heart problems is the electrocardiogram (ECG). It’s essential to accurately categorize ECG data in order to monitor and diagnose a variety of heart conditions. This work suggests a method that effectively classifies arrhythmias using ECG data utilizing deep learning algorithms. The system takes into account the ECG signals that fall into the following four categories are supraventricular premature beat (S), premature ventricular contraction (V), fusion beats (F), and unclassifiable beat (Q). The system surpasses current state-of-the-art techniques by employing an intuitive convolutional neural network (CNN) architecture called “RythmNet,” which results in an overall accuracy of 97.1%. The findings show that the suggested method has the ability to reliably and efficiently classify arrhythmias, which can be extremely important in the detection and follow-up of heart disorders.