Modified Imperialist Competitive Algorithm (MICA) For Smart Heart Disease Prediction in IoT System

Abstract

For the detection and prognosis of heart disease, Internet of Medical Things (IoMT) technology has recently been implemented in healthcare systems. The intended study's main objective is to foresee heart illness using medical data and imaging to classify data. Preprocessing is done on the input dataset to deal with missing values and incorrect data. IoT devices analyse the data they receive from patients, physicians, or nurses using the Modified Imperialist Competitive Algorithm (MICA). The IoT device's analysis of the data allows for effective and informed judgements to be made by humans, robots, and even other IoT devices. A modified imperialist competitive algorithm is suggested in this research in order to pinpoint the essential characteristics of heart disease. The Modified Imperialist Competitive Algorithm is used to select features for the diagnosis of heart disease (MICA). The improved self-adaptive Bayesian algorithm (ISABA) technique is then used to classify the chosen features into normal and abnormal states. For detecting normal sensor data and abnormal sensor data, respectively, the ISABA approach achieved accuracy of 96.85% and 98.31%. With a 96.32% specificity and a 99.15% maximum accuracy in categorizing images, the proposed model outperformed the competition