Fast and non-invasive serum detection technology based on surface-enhanced Raman spectroscopy and multivariate statistical analysis for liver disease

The existing detection methods have the disadvantages of long detection period and high cost for liver disease. In this study, we provide an auxiliary tool for the rapid and sensitive early diagnosis of liver disease. We compared the surface-enhanced Raman spectroscopy of 736 serum samples which contain a variety of hepatitis B related liver disease and analyzed them with orthogonal partial least squares discriminant analysis, and we comprehensively evaluated models through the receiver operating characteristic curve and ten-fold cross validation. The area under curves are all greater than 0.97 and the accuracy rate of the test set reached as high as 95.33%, which show that this method has a high classification accuracy rate for liver diseases. This method is helpful for early diagnosis of liver disease.155