Using Fourier Transform Infrared (FTIR) Spectroscopy Coupled with Multivariate Analysis to Diagnose Polycystic Ovary Syndrome (PCOS)

Abstract

The biochemical diagnostic test is important in the identification of patients with polycystic ovary syndrome (PCOS). Fourier Transfer Infrared (FTIR) measurement which analyzes the overall metabolic profiling of serum may provide more biochemical information. Here we applied the FTIR spectroscopy with multivariate analysis for a PCOS diagnosis. Sixty patients and thirty one healthy subjects were recruited in the present study. FTIR measurement was carried out on the subjects’ serum. The most significant variables obtained from FTIR data were nominated by variable importance in the projection (VIP) value following an OPLS-DA model creating from three different spectral regions: first region/ 900-1200 cm-1, second region/ 1500-1700 cm-1, and third region/ 2800-3100 cm-1. An OPLS-DA of the FTIR data showed variances in the components of patients’ serum. All OPLS-DA models showed validation (P-value < 0.05) based on CV- ANOVA, 900-1200 cm-1 region (R2Y(cum)= 0.926; Q2(cum)= 0.57), 1500-1700 cm-1 (R2Y(cum)= 0.828; Q2(cum)= 0.594), and 2800-3100 cm-1 (R2Y(cum)= 0.927; Q2(cum)= 0.717). These results showed that spectral biomarkers separated patients’ serum. In summary, our work proves that vibrational spectroscopy (FTIR) combined with multivariate analysis (OPLS-DA) showing the possibility to become a chemical reagent-free approach for biochemical assessment of bio-fluid.