Classification of ECG Signal for Cardiac Arrhythmia Detection Using GAN Method

Today, a big number of people suffer from various cardiac problems all over the world. As a result, knowing how the ECG signal works is critical for recognising a number of heart diseases. The electrocardiogram (ECG) is a test that determines the electrical strength of the heart. In an ECG signal, PQRST waves are a group of waves that make up a cardiac cycle. The amplitude and time intervals of PQRST waves are determined for the learning of ECG signals in the attribute removal of ECG signals. The amplitudes and time intervals of the PQRST segment can be used to determine the appropriate operation of the human heart. The majority of approaches and studies for analysing the ECG signal have been created in recent years. Wavelet transform, support vector machines, genetic algorithm, artificial neural networks, fuzzy logic methods and other principal component analysis are used in the majority of the systems. In this paper, the methodologies of support vector regression, kernel principal component analysis, general sparse neural network and   generative adversarial network are compared. The GAN method outperforms both of the other methods. However, each of the tactics and strategies listed above has its own set of benefits and drawbacks. MATLAB software was used to create the proposed system. The proposed technique is demonstrated in this study with the use of the MIT-BIH arrhythmia record, which was used to manually annotate and establish validation.