Artificial Neural Networks for Brain Cancer Classification

Abstract

The detection and categorization of brain tumours is now possible because to new technology. The MRI images of individuals who have been diagnosed with Astrocytoma are analyzed by the system, which makes use of an artificial neural network, in order to identify tumour blocks or lesions also to classify the type of tumour. These approaches are described below. The Grey Level Co-occurrence Matrix (GLCM) was the tool that made it feasible to characterize the different textures of tumours. A contrast is drawn between each of these traits and those that are already included in the Knowledge Base. Finally, after years of research and development, a Neuro Fuzzy Classifier has been created that is capable of correctly identifying the various subtypes of brain cancer. An initial phase of testing focused on learning and training, followed by a second phase focused on recognizing and assessing, have been carried out on the system in its entirety. The Radiology Department at Tata Memorial Hospital (TMH) was kind enough towards supply the training data for the system, which consisted of MRI scans of patients who had already been diagnosed with brain cancer. The method was also validated by using unlabeled MRI image data taken from patients with brain tumours treated at TMH. The classification of these samples and the handling of outliers were both determined to be successfully handled by the system.