Operational research techniques in medical treatment and diagnosis: A review

Abstract

Many modern techniques for the diagnosis of pathological states in humans and for their subsequent treatment can be posed as nonlinear identification problems of essentially nonlinear dynamic systems or as nonlinear optimal control problems. It can be shown that the linearised versions of such models are inadequate and do not represent at all well the complexity of the problem. Thus, nonlinear estimation and control techniques are required for progress to be made in this field. The aim of this review is to examine some models suggested in the medical literature for the modelling of certain medical treatments and diagnoses. Then examine how these models can be enriched by using Operational Research techniques so that a better control is provided on the diagnosis and the treatment, as well as the formulation of more precise models of the phenomenon. The review will present some applications both therapeutic and diagnostic that have appeared in the literature. Special interest will be bestowed on hyperthermic systems in oncological treatment and glucose–insulin dynamics for diabetic patients, while heart dynamics and magnetic resonance imaging will also receive attention. These applications are good examples to show the advantages of Operational Research methods in this field of endeavour. The outline of the paper is the following. After the introduction, in section two a brief description of nonlinear system models of phenomena will be given, for definitional and descriptive purposes. In section three a discussion of how to apply System theory in the medical field will be presented, together with an analysis of the possible benefits. In section four some applications of dynamical models to medical diagnosis and treatment will be described, while in section five the appropriate conclusions will be stated.