Multi-modal spectroscopy of breast tissue
Author(s)
Volynskaya, Zoya I![Thumbnail](/bitstream/handle/1721.1/34121/67766434-MIT.pdf.jpg?sequence=5&isAllowed=y)
DownloadFull printable version (8.952Mb)
Other Contributors
Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
Advisor
Michael S. Feld.
Terms of use
Metadata
Show full item recordAbstract
Breast cancer is the most common form of cancer afflicting women in the United States; one out of eight women will be diagnosed with breast cancer during her lifetime. Currently, screening is performed by a combination of annual clinical breast examinations and x-ray mammography. However, only 10 to 25 percent of suspicious lesions detected during mammography are diagnosed as malignant upon biopsy, which implies that a large number of biopsies can be avoided. Although mammography images anatomic changes, it is not sensitive to the underlying morphological and biochemical changes that distinguish benign and malignant breast lesions. Presently employed diagnostic procedures are invasive, time consuming, and expensive. Thus, there is a clinical need to develop new tools for the early diagnosis of malignancy in the breast. In recent years our laboratory has explored the use of Raman spectroscopy for diagnosing disease; one important area is the detection of breast cancer. Raman spectroscopy provides information about the morphological and biochemical make up of tissue and, with the aid of our diagnostic algorithm, has provided good results in distinguishing between malignant and benign breast lesions, with a sensitivity, specificity, and an overall accuracy of 90, 96, and 86 percent, respectively [Haka,2004]. (cont.) Although these initial results are promising, we would like to improve the overall accuracy. Another promising spectroscopic technique developed in our laboratory is tri-modal spectroscopy (TMS), the combination of diffuse reflectance (DRS), intrinsic fluorescence (IFS), and light scattering spectroscopy (LSS). This technique has been successfully applied to the diagnosis of epithelial neoplastic tissue, leading to the interest in exploring its application to the diagnosis of lesions in breast tissue. Finally, the Raman and DRS/IFS modalities provide complementary information and the combination of this information into a single diagnostic algorithm may provide superior diagnostic capabilities. The central theme of this research is to investigate DRS/IFS as a useful technique for the diagnosis of breast cancer and to evaluate the effectiveness of its combination with Raman spectroscopy. Through this research, we hope to aid the medical community in early diagnosis, treatment, and prevention of breast cancer.
Description
Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2005. Includes bibliographical references (leaves 64-66).
Date issued
2005Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer SciencePublisher
Massachusetts Institute of Technology
Keywords
Electrical Engineering and Computer Science.