Abstract

The aim of this study was to find the differences and relationships between normal, atypical ductal hyperplasia (ADH), ductal carcinoma in situ (DCIS), and invasive ductal carcinoma (IDC) lesions of the breast based on biochemical characteristics determined by Raman spectroscopy (RS). After collecting 39 frozen sections from patients who underwent surgical resection or mammotome biopsy, nine normal tissues, seven ADH, eight DCIS, and 15 IDC lesions were detected using confocal RS. We then used leave-one-out cross-validation (LOOCV) and radial basis function (RBF) to build a support vector machine (SVM) diagnosis model. Pronounced mean Raman spectra differences were observed between normal tissues, ADH, DCIS, and IDC tissues. Most noticeable was the increased protein and reduced lipid levels of ADH tissues compared to normal tissues. The major spectra differences in ADH, DCIS, and IDC spectrograms were evidenced by a red shift with a broad peak of CH2 (1301 cm⁻¹), the intensity of the stretching vibration peak of carotenoids (1526 cm⁻¹), a relatively strong band of amide-I (1656 cm⁻¹), and the nuclear (882 cm⁻¹) acid peak. Atypical ductal hyperplasia tissues had the largest constituent variations between subjects. During the disease progression, IDC tissues have smaller inter-subject constituent variations than DCIS and ADH tissues. The overall accuracy of SVM model is 74.39%. The sensitivities of normal tissue, ADH, DCIS, and IDC are 62.5%, 50%, 90%, and 66.7%, respectively. The specificities of normal tissue, ADH, DCIS, and IDC are 100%, 100%, 66.7%, and 89.06%, respectively. Atypical ductal hyperplasia shows significant differences and the relationship between normal tissue and malignant disease. Further study to explain the biochemical relationships between these differences will shed more light into a better understanding of the mechanism by which ADH converts to DCIS and to IDC.

© 2016 The Author(s)

Automatic recognition of breast invasive ductal carcinoma based on terahertz spectroscopy with wavelet packet transform and machine learning

Wenquan Liu, Rui Zhang, Yu Ling, Hongping Tang, Rongbin She, Guanglu Wei, Xiaojing Gong, and Yuanfu Lu
Biomed. Opt. Express 11(2) 971-981 (2020)

Differential diagnosis of breast cancer using quantitative, label-free and molecular vibrational imaging

Yaliang Yang, Fuhai Li, Liang Gao, Zhiyong Wang, Michael J. Thrall, Steven S. Shen, Kelvin K. Wong, and Stephen T. C. Wong
Biomed. Opt. Express 2(8) 2160-2174 (2011)

Quantitatively characterizing the microstructural features of breast ductal carcinoma tissues in different progression stages by Mueller matrix microscope

Yang Dong, Ji Qi, Honghui He, Chao He, Shaoxiong Liu, Jian Wu, Daniel S. Elson, and Hui Ma
Biomed. Opt. Express 8(8) 3643-3655 (2017)

Scanning point terahertz source microscopy of unstained comedo ductal carcinoma in situ

Kosuke Okada, Quentin Cassar, Hironaru Murakami, Gaëtan MacGrogan, Jean-Paul Guillet, Patrick Mounaix, Masayoshi Tonouchi, and Kazunori Serita
Opt. Continuum 1(3) 527-537 (2022)

Raman difference spectroscopy: a non-invasive method for identification of oral squamous cell carcinoma

Christian Knipfer, Johanna Motz, Werner Adler, Kathrin Brunner, Medhaine Tesfay Gebrekidan, Robert Hankel, Abbas Agaimy, Stefan Will, Andreas Braeuer, Friedrich Wilhelm Neukam, and Florian Stelzle
Biomed. Opt. Express 5(9) 3252-3265 (2014)