A Study on Prediction of Myocardial Infarction Using Computational Intelligence and Machine Learning Algorithms

One of the prominent causes of mortality and disability globally is myocardial infarction (MI). Early and accurate detection of MI is crucial to reduce the risks associated with it. Machine Learning (ML) approaches coupled with Computationally Intelligent Algorithms (CIA) have proved successful in many domains. In this work, an attempt has been made to predict and classify MI using ML techniques and advanced CIA for feature selection. Dataset comprising of patient records from Krasnoyarsk Inter district Clinical Hospital, Russia has been used to classify 12 different MI complications. To choose standout attributes for classification, four swarm intelligence-based algorithms have been used: Particle Swarm Optimization (PSO), Firefly Algorithm (FA), Artificial Bee Algorithm (ABA), and Bat algorithm (BA). For classification, three distinct classifier models—Random Forest (RF), Decision Tree (DT), and K Nearest Neighbor (KNN)— have been developed. In order to better understand the condition and contribute to the creation of novel diagnostic and preventative strategies, the study intends to discover and assess the efficiency of various CIA and ML algorithms in correctly predicting MI.