Portable ECG Device for Remote Monitoring and Detection of Onset of Arrhythmia

The classification of ECG at an early stage of abnormality is vital, since failure to do so increases the risk of a cardiac arrest in real time. The detection of anomalies in the ECG signal helps with the diagnosis of the patient's cardiac status. In this paper a prototype is proposed which obtains the patient's ECG data and classifies it into various classes of arrhythmia in real time. The prototype consists of ECG sensor and Raspberry Pi. The ECG sensor collects patient's data in real time and the collected data is processed on the Raspberry Pi using DWT for ECG feature extraction. The extracted features are given as an input to a trained SVM algorithm to detect the onset of arrhythmia, in order to notify for immediate medical aid. This paper proposes a prototype which processes raw ECG data and detects the onset of arrhythmia by leveraging a trained machine learning algorithm.