AUTOMATED SEARCH FOR PATTERNS OF DISEASE IN INFRARED SPECTRA OF BODY FLUIDS By

A major problem arising from the traditional analysis of the infrared (IR) spectrum of body fluid is its labor intensity and subjectivity. A Fuzzy C-Means Cluster Adaptive Wavelet (FCMC-AW) Algorithm has been proposed to automatically search the patterns of disease by classifying the infrared spectra of body fluids, such as Synovial Fluid (SF) and Cerebrospinal Fluid (CSF). This algorithm consists of a Bayesian Classifier and an Adaptive Wavelet M-band Feature Extractor. The classifier assigns a new pattern to a predefined class in which it acquires the highest posterior probability calculated from the membership matrix of FCMC. The classifier also produces the measure of FCMC-CVQPM. The adaptive capability of the feature extractor comes from the parameterization of the wavelet matrix A(m,g;R) based on the theory of Pollen product. During the training phase of the algorithm, the Quasi-Newton optimizer is employed to find the optimal wavelet matrix A when the measure of FCMC-CVQPM reaches its maximum. Then the feature vector of a new pattern will be extracted by the wavelet decomposition algorithm and subsequently classified by the Bayesian Classifier. The experiments demonstrated that the low frequency components of the signal gained more credits in spectrum classification and that FCMC-AW algorithm could provide additional information on classification confidence by membership matrix. The experiments also demonstrated that an adaptive wavelet had far more potential than a fixed wavelet for the purpose of classification. The experiments of SF and CSF classification revealed that the FCMC-AW algorithm with some certain settings of parameters could be specifically suitable for correlation between SF IR spectra and the variants of arthritis. ii PERMISSION TO USE In presenting this thesis in partial fulfillment of the requirements for a Master of Science degree from the University of Saskatchewan, I agree that the Libraries of this University may make it freely available for inspection. I further agree that permission for copying of this thesis in any manner, in whole or in part, for scholarly purposes may be granted by the professor or professors who supervised my thesis work or, in their absence, by the Head of the Department or the Dean of the College in which my thesis work was done. It is understood that any copying or publication or use of this thesis or parts thereof for financial gain shall not be allowed without my written permission. It is also understood that due recognition shall be given to me and to …