Analysis on Detecting Cardiac Arrhythmia Using Advanced Technologies

Essential features of the various cardiac diseases that affect the human heart are provided by the electrocardiogram (ECG). A significant portion of the diagnosis of heart illness comes from the categorization of arrhythmias. An arrhythmia is any alteration to the regular electrical impulse sequence. Conventional techniques for analyzing digital signals, machine learning, and the offshoots, including deep learning approaches are widely used for analyzing and classification of ECG signals and more importantly, for the timely diagnosis of cardiac disorders and arrhythmias. An extensive review of the literature regarding ECG analysis is presented in this document. Also address the computational barriers of machine learning techniques to wearable devices. The study started with a search for pertinent papers and then the gathered data was examined. Examining the assessed ECG features and the machine learning technique for identifying arrhythmias was the second step. The analysis shows that even though the MIT-BIH database requires a large amount of preprocessing work, it chose a sizable proportion of studies. We examine the substantial corpus of literature on pre-recorded ECG data, signal collection, ECG signal processing, and noise reduction, identifying ECG spectrographic states through function technology, classification of ECG signals, and relative perspectives among the examined modules. We reviewed the ultra-modern machine learning strategies for automated arrhythmia detection.