3-D MRI Brain Scan Classification of Epilepsy versus Non-Epilepsy

A 3-D classification method is presented for Magnetic Resonance Imaging (MRI) brain scan volumes of interest. The objective of the classification is to identify volumes that feature indicators of epilepsy against volumes that do not by considering the shape and size of the lateral (left and right) ventricles of the brain. The dataset used with respect to the reported experimentation comprised 210 3-D MRI brain scans of which 105 were from epilepsy patients and the remainder from healthy people. The classification is supported by two proposed point-series based representation techniques: (i) Disc and (ii) Spoke. The reported experimental results indicated that our proposed techniques can classify 3-D MRI brain scan data with a classification accuracy of up to 69.81%.