Medical model estimation with particle swarm optimization

Abstract

In this paper, a nonlinear medical model based on observational variables has been produced and the particle swarm optimization (PSO) technique, which is an effective technique to predict optimum parameters of the biomedical model, has been used. This study has been conducted on a dataset consisting of 539 subjects. For comparison purposes, nonlinear regression analysis, nonlinear deep learning, and nonlinear regression neural network methods are also considered and the PSO results appear to be slightly better than that of other methods. Built on observational variables and findings, the model is expected to be a good guide for healthcare professionals in diagnosing pathologies and planning treatment programs for their patients. It is therefore strongly believed that the article will be particularly useful for those interested in emerging biomedical models in various medical modelling areas such as infectious and hematological diseases such as anemia.