Computer assisted impeccable examining of brain tumor classification and feature extraction in mri images using multi modal features and machine learning methods

Abstract

Cancer is the second leading cause of death for both men and women in worldwide and is expected to become the leading cause of death in the next several decades It has been shown that early detection and treatment of brain cancer are the most effective methods of reducing mortality The rapid development in image processing and soft computing technologies have greatly enhanced the interpretation of medical images and contributed to early diagnosis This accounts for 13 of all deaths for that year making cancer a common threat to all families As technology becomes more efficient a trend towards computer aided diagnostic CAD tools for identification prognosis prediction and reoccurrence likelihood is becoming a reality A cornerstone of CAD systems in the field of digital histopathology where analyzing large quantities of cellular images can leverage the research from the mature field of computer vision In this thesis three novel methods were developed for automatic detection and classification of brain tumour in MRI images Image texture can deliver important information on the abnormality of tissue and this thesis capitalizes on this knowledge of tumour texture grading and classification The main goal of this thesis is to design implement and evaluate a pattern recognition system to aid in brain tumour classification accuracy by analysing regularly taken T1 post contrast MRI images In the first work a novel brain tumor classification approach in MRI images using MMTF with LS SVM has been developed Two major contributions of this research are feature extraction and classification In feature extraction Multi model texture features is used to extract various texture and intensity based feature using GLCM Wavelets and Gabor Filters newline