Diagnosing diabetes mellitus from human breath odor using by QCM-SSC gas seansor array

DIAGNOSING DIABETES MELLITUS FROM HUMAN BREATH ODORUSING BY QCM-SSC GAS SENSOR ARRAYNazmi ÇAKMAKElectrical and Electronics Engineering MSc. Thesis, 2007Supervisor: Assist. Prof. Dr. Hamdi Melih SARAOĞLUSUMMARYVarious parameters produced by a human body can be used for diagnosing illnesses.For example, blood, urine and gaita are very common and used frequently in diagnosis. It isknown that skin odor and breath odor, as the other parameters produced by a human body carryvaluable information about the healthiness. In this perspective, it is considered that someillnesses can be diagnosed by examining the breath odor.It considered that diabetes can be diagnosed from human breath using an electronic nosewhich consists of a QCM sensor array. In this thesis, the QCM based E-nose, developed fordiagnosing diabetes is introduced and experimental results are presented. Experimental resultsand real measurements of blood glucose are compared using PCA (Principal ComponentAnalysis) and ANN (Artificial Neural Networks) techniques. Moreover, influence of breathhumidity over QCM sensors have been examined using same techniques.The work has been conducted in the scope of TUBITAK Project, No: 104E053:?Diagnosing System Design for Medical Applications Using by QCM-SSC Gas Sensor Array?.Keywords: Breath Odor, Diabetes, Electronic Nose, QCM Sensor.