Exploring the Feasibilities of Applying Min-Max Threshold Analysis with Machine Learning Techniques for Categorization of X-Wave in ECG Signal

Abstract

A non-stationary signal called an electrocardiogram (ECG) is used to assess the rhythm and tempo of a person's heartbeat. Feature extraction is the primary phase in ECG classification, since it is responsible for identifying a group of pertinent characteristics that can achieve the greatest accuracy. After everything is said and done, this study provides a comprehensive overview of the methods currently used for detecting ECG waveforms. This study compares and contrast the current methods for ECG classification and ECG waveform detection and highlight their respective strength and weakness. The major goal of this study is to offer an automated ECG wave identification and classification method. From the outcomes, it can be decided as the accuracy is need to be enhanced/improved. The X-wave of ECG could be recognized using Min Max threshold analysis method. Then it is subjected to classification by means of Convolutional Neural Network (CNN). It is anticipated that this evaluation will prove to be an efficient tool for researchers, scientific engineers, and others engaged in this field to identify pertinent sources.