Classification/Diagnosis of Brain Tumors Using Discriminant Function Analysis

Abstract

The occurrence of brain tumors is one of the most important causes of morbidity and mortality in young adults and children. The accurate classification of a tumor is crucial for finding the best treatment of the patient. Therefore, the development of diagnostic methods that may improve classification of brain tumors is of grate importance. In this work, a review of current method applied to the diagnosis of brain tumors is presented. Special attention is paid to the remarkable opportunities provided by using discriminant function analysis as a tool that may support classification of brain tumors. The examples of applications of selected techniques of modern physics such us magnetic resonance imaging, magnetic resonance spectroscopy, synchrotron radiation based X-ray fluorescence, infrared spectroscopy and small angle X-ray scattering technique for classification of brain tumors are discussed. Moreover, the usefulness of morphological features of neoplastic tissues including blood vessel shape and morphometric features of cell nuclei for diagnosis of brain tumors with the use of discriminant function analysis is presented. As the literature studies show, very high predictive abilities are achieved if the analytical and imaging results are supported by discriminant function analysis. Therefore, the potential diagnostic methods in combination with discriminant function analysis would be recommended to be included to standard clinical examinations.