Elemental analysis and identification of papillary thyroid cancer tissues using laser-induced breakdown spectroscopy
Min Luo,Xiaohui Li,Qiuxia He,Jing Liu,Hansheng Lu,Qi Liu,Xiuhua Yang
DOI: https://doi.org/10.1039/d1ja00367d
2022-01-01
Journal of Analytical Atomic Spectrometry
Abstract:Thyroid function depends on the presence of certain elements necessary for both synthesis and metabolism of thyroid hormones. Deficiency or excess of certain elements may trigger biological abnormalities and cause thyroid diseases and even cancers. In this work, elemental analysis and identification of papillary thyroid cancer (PTC) tissues were investigated using laser-induced breakdown spectroscopy (LIBS). The PTC (n = 14) and normal (n = 13) thyroid tissue samples were obtained from PTC patients undergoing surgical resection of the thyroid gland. The biopsies were cut into 30 mu m-thick slices from wax blocks, fixed on silicon slides, and ablated using a 1064 nm Q-switched Nd:YAG laser. Ten spectral emissions of calcium (Ca), magnesium (Mg), sodium (Na), potassium (K), carbon (C), oxygen (O), nitrogen (N), hydrogen (H), and molecular CN were compared between the PTC and normal tissues. This shows that Ca and Mg levels in the PTC tissues are significantly higher (p < 0.005) than those in the normal ones, suggesting that Ca and Mg may play important roles in the development and metabolic process of PTC. The ten emissions were further used to discriminate the PTC tissues from the normal ones, in combination with several chemometric methods, including principal component analysis (PCA), linear discriminant analysis (LDA) classifier, and support vector machine (SVM) classifier. Leave-one-out cross-validation shows that the PTC tissues can be discriminated from the normal ones. Total accuracies of 92.6% and 81.5% were obtained by the SVM classifier and LDA classifier, respectively. The corresponding sensitivity values were 0.929 and 0.714, respectively, and the specificity values were both 0.923. The results show that LIBS coupled with chemometric methods may serve as a potential tool for detection of papillary thyroid cancer.