0
Research Papers

Diagnosis of breast cancer using fluorescence and diffuse reflectance spectroscopy: a Monte-Carlo-model-based approach

[+] Author Affiliations
Changfang Zhu

University of Wisconsin-Madison, Department of Electrical and Computer Engineering, Madison, Wisconsin 53705

Gregory M. Palmer

Duke University, Department of Radiation Oncology, Durham, North Carolina 27710

Tara M. Breslin

University of Michigan, Division of Surgical Oncology, Ann Arbor, Michigan 48109

Josephine Harter

University of Wisconsin, Department of Pathology, Madison, Wisconsin 53705

Nirmala Ramanujam

Duke University, Department of Biomedical Engineering, Durham, North Carolina 27708

J. Biomed. Opt. 13(3), 034015 (May 31, 2007December 16, 2007January 11, 2008May 30, 2008). doi:10.1117/1.2931078
History: Received May 31, 2007; Revised December 16, 2007; Accepted January 11, 2008; Published May 30, 2008
Text Size: A A A

We explore the use of Monte-Carlo-model-based approaches for the analysis of fluorescence and diffuse reflectance spectra measured ex vivo from breast tissues. These models are used to extract the absorption, scattering, and fluorescence properties of malignant and nonmalignant tissues and to diagnose breast cancer based on these intrinsic tissue properties. Absorption and scattering properties, including β-carotene concentration, total hemoglobin concentration, hemoglobin saturation, and the mean reduced scattering coefficient are derived from diffuse reflectance spectra using a previously developed Monte Carlo model of diffuse reflectance. A Monte Carlo model of fluorescence described in an earlier manuscript was employed to retrieve the intrinsic fluorescence spectra. The intrinsic fluorescence spectra were decomposed into several contributing components, which we attribute to endogenous fluorophores that may present in breast tissues including collagen, NADH, and retinol/vitamin A. The model-based approaches removes any dependency on the instrument and probe geometry. The relative fluorescence contributions of individual fluorescing components, as well as β-carotene concentration, hemoglobin saturation, and the mean reduced scattering coefficient display statistically significant differences between malignant and adipose breast tissues. The hemoglobin saturation and the reduced scattering coefficient display statistically significant differences between malignant and fibrous/benign breast tissues. A linear support vector machine classification using (1) fluorescence properties alone, (2) absorption and scattering properties alone, and (3) the combination of all tissue properties achieves comparable classification accuracies of 81 to 84% in sensitivity and 75 to 89% in specificity for discriminating malignant from nonmalignant breast tissues, suggesting each set of tissue properties are diagnostically useful for the discrimination of breast malignancy.

Figures in this Article
© 2008 Society of Photo-Optical Instrumentation Engineers

Citation

Changfang Zhu ; Gregory M. Palmer ; Tara M. Breslin ; Josephine Harter and Nirmala Ramanujam
"Diagnosis of breast cancer using fluorescence and diffuse reflectance spectroscopy: a Monte-Carlo-model-based approach", J. Biomed. Opt. 13(3), 034015 (May 31, 2007December 16, 2007January 11, 2008May 30, 2008). ; http://dx.doi.org/10.1117/1.2931078


Access This Article
Sign In to Access Full Content
Please Wait... Processing your request... Please Wait.
Sign in or Create a personal account to Buy this article ($20 for members, $25 for non-members).
 
Your Session has timed out. Please sign back in to continue.
Sign In to Access Full Content

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Related Book Chapters

Topic Collections

PubMed Articles
Advertisement

Buy this article ($18 for members, $25 for non-members).
Sign In