Brain Tumors Classification from MR Images Using a Neural Network and the Central Moments

For the medical images classification, we proposed a novel method for the classification of Magnetic Resonance (MR) images of the brain. The proposed system is based on the use of a new method for the features extraction. The principal objective is to calculate the histogram of each zone selected by sliding a window of size 16×16 Pixels on the MR image of the brain, this allow us to obtain sixty four (64) histogram, and each obtained histogram will be considered as a sequence for which we calculate the central moments of order 1, 2 and 3. The classification is achieved by a multi-layer perception.