Classi ication of Electrocardiogram (ECG) signals plays a signi icant role in the identi ication of the functioning of the heart. This work pertains with the ECG signals, where the classi ier is developed for identi ication of normal or abnormal conditions of the heart. The raw ECG signals are collected from an online database (www.physioNet.org) for classi ication. The raw ECG signal is pre-processed for noise removal, and the frequency spectrum is analysed to compare raw and denoised ECG signal. Attributes (P, Q, R, S, T time intervals) from denoised ECG signal is analysed and classi ied using Convolution Neural Network (CNN). The paper reports a classi ication technique to differentiate ECG signals from the MIT-BIH database (arrhythmia database, arrhythmia pwave annotations, atrial ibrillation). The CNN analyses the deviation between nominal ranges of attributes (amplitude and time interval) and classi ies between the abnormality and normal ECG wave. This work provides a simple method for interpreting ECG related condition for the clinician and helps medical practitioners to make diagnostic decisions.

In this paper, we propose a deep learning convolutional neural network (CNN) approach to classify arrhythmia based on the time interval of the QRS complex of the ECG signal. The ECG signal was denoised using multiple filters based on the Pan Tompkins algorithm. QRS detection has been done using Pan Tompkins Algorithm. Then the QRS complex is identified using local peaks based technique inside the layers of the Convolutional Neural Network where the repeated application of the same filter to our input results in a map of activations called a feature map, indicating the locations and strength of a detected feature in an input which in our case is the changes in the q-s time interval. Based on the R-R time interval, Heart rate variability (HRV) was computed, and Poincare plot was generated. Instead of using raw ECG signal to train the CNN, we used the feature extracted from ECG signal obtained from Physionet database to train the CNN and map the pattern changes for different classes of diseases. The classifier was then used to classify the test input as either or normal, tachyarrhythmia or intracardiac atrial fibrillation. Data acquisition, ECG data pre-processing and CNN classifier are the several methods that are involved for the classification of several arrhythmias.

Analysis of Electrocardiogram (ECG) signal can lead to better detection of cardiac arrhythmia. The important steps involved in the ECG signal analysis include acquisition of data, pre-processing of signal to remove artefacts, feature extraction of attributes and finally identifying abnormalities. This work proposes an efficient implementation of the R-R interval-based ECG classification technique for detecting abnormalities in heart functioning. ECG signals from an online database (PhysioNet.org) was analysed after noise removal for R-R interval, as R peak has the maximum prominent amplitude in ECG wave. Deviation in the R-R interval values obtained from unhealthy was observed and compared with healthy subjects. This observation of cardiac activity can be visualised in our developed Graphical User Interface (GUI). The GUI platform requires only the input of the ECG signal that is to be analysed for abnormalities, which can provide the clinician with the result of cardiac abnormality classification and can help in diagnosis.