ECG Classification Using Deep Neural Networks

ECG Classification Using Deep Neural Networks

Author Name : Sumit Singh Rajput, Shivam Makade, Shubham Potdukhe, Sumit Morkhade, Prof. Anand Sharma

ABSTRACTION­­­

Electrocardiogram (ECG) is an important diagnostic tool for analysing cardiac tissues and structures. It includes information about the heart structure and the function of its electrical conduction system.Time series classification (TSC) is the area of machine learning interested in learning how to assign labels to time series.Now a days, Deep learning has received enormous attention because of its high scalability.The accuracy of deep learning for TSC has been raised to a competitive level. In our study, we developed different Neural Network which are used in TSC. We used simple CNN, LSTM, GRU and Inception Time out of which InceptionTime showed great results. Inception Time uses power of CNN in efficient way With Inception Time we got about 98.89 percent of accuracy. We also cross checked our results with doctor and the results were matching with his diagnosis too.

Keywords: Deep Learning, TSC (Time Series Classification), Imbalanced Data, ECG, DNN