A review of methods for myocardial infarction detection using of electrocardiographic features

Abstract

Various kinds of methods have been developed to detect the myocardial infarction (MI) which becomes one of the biggest causes of death. In this case, early detection of MI is necessary to be done. This article provided the completion of detection techniques for the MI. The detection of MI can be done by using 12 signals of the EKG by submitting entropy energy and morphological features. The initial stages may consist of data preprocessing to eliminate noise and down-sampling, pulse segmentation, data augmentation, and QRS detection. This stage was done to find the electrocardiographic features that can give feature of the incidence of MI. This feature was used for the input methods such as Support Vector Machine and convolutional neural networks (CNN). This study indicated that the detection done using SVM can show performance testing until 99.81%.