ILDIZ, GÜLCE ÖĞRÜÇ2019-09-052019-09-052019https://hdl.handle.net/11413/5242In this study, Fourier-transform infrared spectroscopy complimented with multivariate analysis (principal component analysis and partial least square methods) is used for the first time for the classification of bipolar and schizophrenia disorders. Our main goal was to identify spectral changes correlated with some biomarkers associated with schizophrenia and bipolarity. We used the spectral information for the calibration of statistical models to discriminate and classify blood plasma samples belonging to bipolar and schizophrenic patients. FT-IR spectra of 30 samples of blood plasma obtained from each, bipolar and schizophrenic patients and healthy control group were collected. The results obtained from principal component analysis (PCA) show a clear discrimination between the bipolar (BP), schizophrenic (SZ) and control group’ (CG) blood samples that also give possibility to identify three main regions that show the major differences correlated with both mental disorders (biomarkers). Furthermore, a model for the classification of the blood samples was calibrated using partial least square discriminant analysis (PLS-DA), allowing the correct classification of BP, SZ and CG samples. The results obtained applying this methodology suggests that it can be used as a complimentary diagnostic tool for the detection and discrimination of these mental diseases.en-USAttribution-NonCommercial-NoDerivs 3.0 United Stateshttp://creativecommons.org/licenses/by-nc-nd/3.0/us/conferenceObject