Information system for predicting relapses of myocardial infarction, implemented as a mobile application

Abstract

An important task of improving the quality of medical care for patients with coronary heart disease (CHD) and those who have suffered acute myocardial infarction (MI) is timely and accurate prediction of possible relapses and progression of myocardial infarction in the studied category of patients. The aim of the study is to develop an information system that allows predicting the likelihood of recurrence of myocardial infarction based on the construction of fuzzy mathematical models. Soft computing technology based on the methodology of synthesis of hybrid fuzzy decision rules was chosen as a mathematical apparatus for the development of an information system. To build the model, the method of forming a feature space was implemented. For this purpose, a group of experts consisting of six experts, consisting of doctors specializing in cardiology, was formed. At the expert level, the composition of informative signs was established, consisting of 18 signs, which were divided into four blocks. According to the formed set of features, a fuzzy predictive model was synthesized. The quality of the information system was checked by expert evaluation, mathematical modeling and statistical tests on representative control samples. In the course of the conducted studies, it was shown that the confidence in the correct prognosis exceeds the value of 0.85, which is a good result for medical prognostic tasks. The space of informative features is formed in the work. The implemented methods and the synthesized prognostic model allow us to recommend the results of research in medical practice both as part of information systems for cardiologists and in the form of application programs for tablets and smartphones.