The most important part of the early prevention of atherosclerosis and cardiovascular diseases is the estimation of the cardiometabolic risk (CMR). The CMR estimation can be divided into two phases. The first phase is called primary estimation of CMR (PE-CMR) and includes solely diagnostic methods that are non-invasive, easily-obtained, and low-cost. Since cardiovascular diseases are among the main causes of death in the world, it would be significant for regional health strategies to develop an intelligent software system for PE-CMR that would save time and money by extracting the persons with potentially higher CMR and conducting complete tests only on them. The development of such a software system has few limitations - dataset can be very large, data can not be collected at the same time and the same place (eg. data can be collected at different health institutions) and data of some other region are not applicable since every population has own features. This paper presents a MATLAB solution for PE-CMR based on the ensemble of well-learned artificial neural networks guided by evolutionary algorithm or shortly EANN-EA system. Our solution is suitable for research of CMR in population of some region and its accuracy is above 90 %.
Keywords: Artificial neural network; Cardiometabolic risk; Evolutionary algorithm; Intelligent healthcare.