Intuitionistic Fuzzy Recurrence Plots for Classifying Cardiac Arrhythmias Using Deep Learning

This article proposes a method for the classification of different types of cardiac arrhythmias through a deep learning model that utilizes recurrence graphs constructed through a fuzzy intuitionistic clustering technique. The utilization of recurrence graphs is predicated on their capacity to encapsulate the pertinent information of the ECG signal into a succinct graphical representation, thereby facilitating the analysis and identification of aberrant patterns. To substantiate the efficacy of the proposed method, ten deep learning models with disparate configurations were trained, and the outcomes were contrasted with those of existing methodologies in the literature. The findings demonstrate an exceptional performance, with an accuracy of 98%, underscoring the promise of recurrence graphs and convolutional neural networks in signal analysis.