Raman micro-spectroscopy and multivariate analysis for the differentiation of skin cancers

Skin cancer is a growing health concern worldwide. In this work, confocal Raman micro-spectroscopy (CRM) was combined with two types of multivariate analysis, principal component analysis (PCA) and linear discriminant analysis (LDA), to accurately differentiate between skin cancer and normal skin. CRM was employed to study three distinct intracellular regions – cytoplasm, nucleoplasm, and nucleolus – within human metastatic melanoma (SK) and skin fibroblast (BJ) cells. PCA/LDA was 92-98% successful in discriminating BJ from SK cells, with higher RNA identified in the nucleoli of BJ cells and higher lipids and collagen identified in the cytoplasm of SK cells. CRM measurements were also done on SCC, BCC, and normal skin tissue samples to determine the feasibility of combining Raman spectroscopy with CO2 ablation. Differentiation with PCA was possible between normal and SCC tissue that had been ablated, with 78% correct identification when non-ablated and 92% when ablated.