Raman Hyperspectroscopy and Chemometric Analysis of Blood Serum for Diagnosing Celiac Disease in Adults

Abstract

Celiac disease (CD) is a chronic autoimmune disorder triggered by an abnormal immune response to gluten, a protein found in wheat, barley, and rye. Current diagnostic methods, including serological assessments and biopsies, can be challenging due to the disease’s heterogeneous nature, creating a need for a reliable, noninvasive diagnostic approach. Here, in this study, we aimed to extend the Raman peak area ratios approach to the adult population. However, our findings indicate no significant differences in Raman peak area ratios between healthy and diseased adults based on blood serum samples. Nevertheless, genetic algorithm combined with partial least squares discriminant analysis (GA-PLS-DA) allowed differentiation with 92% sensitivity and 96% specificity at the spectral level in external validation. Receiver operating characteristic (ROC) analysis showed 100% classification at the donor level in external validation. These results demonstrate further that Raman spectroscopy, combined with chemometrics, is a promising, noninvasive tool for CD diagnosis.