EE Seminar: Brain Tumor Classification of Glioblastomas, Brain Metastasis, Meningioma and CNS Lymphoma

~~Speaker: Nir Dvorecki, 
M.Sc. student under the supervision of Prof. Amir Averbuch and Prof. Shai Avidan

Monday, July 20th, 2015 at 15:00
Room 011, Kitot Bldg., Faculty of Engineering

Brain Tumor Classification of Glioblastomas, Brain Metastasis, Meningioma and CNS Lymphoma

Abstract

The objective of this work is to investigate the use of conventional MRI, DTI and Perfusion imaging in a pattern recognition and machine learning framework for the automatic classification of brain tumors. We propose a complete pipeline consisting of bias correction, normalization, feature extraction, feature selection, and classification. Median intensities are extracted from the enhancing, non-enhancing and edema sections. Feature selection is performed using a leave-one-out cross validation technique. We test our model on a dataset consisting of patients with tumors of types Metastasis, Meningioma, Glioblastoma and CNS Lymphoma. We present a hierarchical classifier and analyze its performance.