Molecular Profiling Thyroid Cancer Using 2T2D-FTIR Spectroscopy Integrated With Machine Learning Models

Thyroid cancer is the most common endocrine malignancy. Two-trace two-dimensional Fourier transform infrared spectroscopy (2T2D-FTIR) shows great promise in cancer research by providing detailed molecular information that helps distinguish tumor types and stages. In this study, we used 2T2D-FTIR combined with machine learning (ML) to diagnose thyroid cancer across different population groups. The study population was divided into three groups: G1 (20-40 ± 2 years, female), G2 (45-60 ± 2 years, male and female), and G3 (65-80 ± 2 years, female). This advanced technique enhances the comparison between healthy and malignant tissues by detecting subtle molecular changes in two-dimensional space. Our findings support the use of 2T2D-FTIR combined with ML for early thyroid cancer detection, and possibly for other cancers as well. The study had a limited sample size due to ethical and availability constraints; however, the ML approach improved accuracy and could potentially achieve better results with larger datasets.