Differentiation of cancer cells in 2D and 3D breast cancer models by Raman spectroscopy

Yi Fang, Purdue University


Raman spectroscopy, due to its fast speed, label-free, high sensitivity and minimal background noise from water in aqueous media free from background noise due to water in aqueous media, has been widely used in biological explorations. In this work, by adopting both 2D and 3D models, we demonstrate the applications of Raman spectroscopy for the classification of cancer cells in different stages: from non-malignant, pre-malignant, malignant to metastatic using breast epithelial cancer cell MCF10 series as the model. Principal Component Analysis (PCA) following by Permutation Multivariate Analysis of Variance (PMAV) were introduced to analyze the spectra of cell lines in different stages, showing that cell lines behave are not exactly the same in 2D and 3D models. PCA in conjunction with linear discriminant analysis were also applied to show sensitivity and specificity determination. Diversities in chemical groups relating to nucleic acids, proteins, and lipids, among other chemicals elucidated by the Raman band assignment, account for differences among the four cancer cell lines and stages of the disease.




Irudayaraj, Purdue University.

Subject Area

Biomedical engineering|Medical imaging|Oncology

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server