Classification of Tuberculosis by Surface-Enhanced Raman Spectroscopy (SERS) with Principal Component Analysis (PCA) and Partial Least Squares – Discriminant Analysis (PLS-DA)

Abstract

Surface-enhanced Raman spectroscopy was employed for the characterization and comparison of classes of tuberculosis (TB) including pulmonary and extra-pulmonary TB with healthy serum samples. Notably, two subgroups of SERS spectra were observed due to biochemical features in the SERS peaks for extra-pulmonary TB. These two subgroups are classified as extra-pulmonary TB type 1 and type 2. SERS clearly differentiates classes of tuberculosis infection from healthy controls on the basis of the SERS features. To demonstrate the discrimination of SERS datasets of tuberculosis classes and healthy controls, principal component analysis (PCA) was used to identify the biochemical differences. Moreover, partial least squares – discriminant analysis (PLS-DA) was used for the classification of tuberculosis types and healthy controls. The sensitivity and specificity were 85% and 90%, respectively.