Wavelet transform of breast tissue fluorescence spectra: a technique for diagnosis of tumors

Abstract

Polarized fluorescence spectra of malignant, benign, and normal human breast tissues in the emission range of 500-700 nm, with an excitation wavelength of 488 nm, are analyzed through discrete wavelet transform. The multiresolution and localization properties of the wavelets are shown to be ideally suited for identifying characteristic features distinguishing these tissue types. Analysis of a number of data sets, belonging to both parallel and perpendicular polarized spectra, have led to several key distinctions between different tumors and corresponding normal breast tissues, revealing the usefulness of polarized fluorescence in the diagnosis of tumors. Wavelet transform also naturally leads to the dimensional reduction of the data set, in the form of low-pass coefficients, making it amenable for physical modeling.