FTIR-based spectrum of salivary exosomes coupled with computational-aided discriminating analysis in the diagnosis of oral cancer.

摘要: To determine the Fourier-transform infrared (FTIR) spectra of salivary exosomes from oral cancer (OC) patients and healthy individuals (HI) to assess its diagnostic potential using computational-aided models. Whole saliva samples were collected 21 OC 13 HI. Exosomes pelleted differential centrifugation (12,000g, 120,000g). The mid-infrared (IR) absorbance (900–5000 cm− 1 range) measured MIR8025 Oriel IR equipped with a PIKE MIRacle ZnSe attenuated total reflectance attachment. Machine learning techniques, utilized build discrimination models for data HI, included principal component analysis–linear discriminant analysis (PCA–LDA) support vector machine (SVM) classification. Sensitivity, specificity area under receiver operating characteristic curve calculated. consistently different HI at 1072 cm− 1 (nucleic acids), 2924 cm− 1 2854 cm− 1 (membranous lipids), 1543 cm− 1 (transmembrane proteins). PCA–LDA model correctly classified sensitivity 100%, 89% accuracy 95%, SVM showed training 100% cross-validation 89%. We specific spectral signature exosomes, which was accurately differentiated based on detecting subtle changes in conformations proteins, lipids nucleic acids optimized artificial neural networks small sets. This non-invasive method should be further investigated diagnosis very early stages or lesions malignant transformation.