Assessment of the effectiveness of serum-infrared spectroscopy in conjunction with multivariate analysis methods for atherosclerosis diagnosis.

Atherosclerosis is a progressive disease characterized by lipid accumulation and fibrous elements in large and medium-sized arteries, and remains a leading cause of death worldwide. A deeper understanding of its molecular nature is critical for developing novel strategies for the prevention, diagnosis, and treatment. This study evaluates attenuated total reflectance-Fourier transform infrared (ATR-FTIR) spectroscopy combined with multivariate analysis techniques to differentiate and classify atherosclerosis by identifying disease-specific spectral variations. Spectral analyses indicated statistically significant differences in lipid (p < 0.0001), protein (p < 0.01), nucleic acid (p < 0.0001), and glucose (p < 0.0001) content of serum samples in the atherosclerosis group compared to controls. Patients with atherosclerosis exhibit altered lipid metabolism, marked by a decrease in saturated lipids and an increase in unsaturated lipids compared to healthy individuals. Additionally, elevated levels of protein, RNA, glucose, and conformational changes in DNA were key spectral features, distinguishing atherosclerosis from controls. Principal component analysis (PCA) successfully differentiated patients from controls, while classification models based on linear discriminant analysis (LDA) and support vector machine (SVM) achieved accuracies of 96.61% and 93.22%, respectively. The ability of FTIR spectroscopy to detect subtle biochemical alterations suggests its potential for early diagnosis. These molecular markers may appear prior to clinical symptoms, highlighting the method's potential for future screening, pending validation in at-risk or preclinical cohorts.