Inteligentni softverski sistem za dijagnostiku metaboličkog sindroma

Abstract

Doktorska disertacija razmatra problem algoritamske dijagnostike metaboličkog sindroma na osnovu lako merljivih parametara: pol, starosna dob, indeks telesne mase, odnos obima struka i visine, sistolni i dijastolni krvni pritisak. U istraživanju su primenjene i eksperimentalno ispitane tri različite metode mašinskog učenja: stabla odluke, linearna regresija i veštačke neuronske mreže. Pokazano je da veštačke neuronske mreže daju visok nivo prediktivnih vrednosti dovoljan za primenu u praksi. Korišćenjem dobijenog rezultata definisan je i implementiran inteligentni softverski sistem za dijagnostiku metaboličkog sindroma.

The doctoral dissertation examines the problem of algorithmic diagnostics of the metabolic syndrome based on easily measurable parameters: sex, age, body mass index, waist and height ratio, systolic and diastolic blood pressure. In the study, three different methods of machine learning were applied and experimentally examined: decision trees, linear regression and artificial neural networks. It has been shown that artificial neural networks give a high level of predictive value sufficient to be applied in practice. Using the obtained result, an intelligent software system for the diagnosis of metabolic syndrome has been defined and implemented.