ECG Classification Using Machine Learning on Wave Samples for the Indian Population

Availability of key human parameters located remotely, to a trained medical professional can result in early diagnosis and prognosis of a patient, leading to better healthcare and thereby reducing the patient load in government hospitals. However, the testing of these key parameters requires sophisticated equipment and trained manpower, which requires heavy investment. The setup for ECG measurement in multi- specialty hospitals is quite costly and bulky, involving 12 sensors, which are attached to the patient’s limbs and chest. This study aims to develop machine-learning models, which can identify whether the cardiac rhythm is normal or abnormal directly using the sample values of the ECG waves, using just 3 leads instead of 12, hence reducing the complexity of setup. Secondly, the research aims at comparing the accuracy of several algorithms for this binary classification of rhythm. Lastly, the dataset used during the study has been prepared and processed on the basis of actual data of ECG records of heart patients by visiting various hospitals in Nagpur city.