Abstract
Hepatitis C has become one of the higher health problems around the world. Near-infrared Raman spectroscopy had been used to spectrally differentiate among health human blood serum from the one with hepatitis C contaminationin vitro. In this study a Raman spectrometer with 80 mW, 830 nm excitation, liquid-nitrogen cooled CCD and imaging spectrograph were used to collect Raman scattering from 24 blood samples (14 healthy and 10 diseased) with collection time of 120 s. It has been used an algorithm based on the Principal Components Analysis (PCA) for main spectral features identification and Mahalanobis distance for blood spectrum classification depending on the serology. It was observed that the highest spectral differences between the two types of human blood serum were found in 1002, 1169, 1262 and 1348 cm−1 Raman bands. The spectral analysis using multivariate statistics presented good results when compared to classical diagnosis for viral hepatitis C, showing that Raman spectroscopy can classify human blood serum spectrum in one of the two categories by identifying biochemical alterations that occur in the presence of viral infections.