Risk Quantification of Metabolic Syndrome with Quantum Particle Swarm Optimisation

Metabolic syndrome (MetS) is a combination of interrelated risk factors associated with an increased risk of developing type II diabetes Mellitus (T2DM), stroke and cardiovascular diseases (CVD). The economic, social and medical burden coupled with increased morbidity of the aforementioned diseases makes their prevention an active research area. Currently, the traditional method of MetS diagnosis is based on dichotomised definitions provided by various expert health organisations. However, this method is laced with the indetermination of MetS in individuals with borderline risk factor values due to a binary diagnosis and the assumption of equal weighting for all risk factors during diagnosis. The purpose of this paper is to examine the use of the MetS areal similarity degree risk analysis based on weighted radar charts comprising of diagnostic thresholds and risk factor results of an individual. We further enhance this risk quantification method by applying quantum particle swarm optimization to derive the weights. The proposed risk quantification was carried out using a sample of 528 individuals from an examination survey conducted between 2007 and 2014 in Serbia. The results are evaluated with the traditional dichotomised method of MetS diagnosis, in this case the joint interim statement (JIS). The results obtained showed that the proposed risk quantification method outperformed the dichotomised method at diagnosing MetS even in individuals who present risk factor examination values at the threshold borderlines.