Detection of Brain Tumors Based on Automatic Symmetry Analysis

This article focuses on the detection of abrain tumor location in magnetic resonance images.The aim of this work is not the precise segmentationof the tumor and its parts but only the detection ofits approximate location. It will be used in futurework for more accurate segmentation. For this rea-son, it also does not deal with detecting of the im-ages containing the tumor. The algorithm expects a2D T2-weighted magnetic resonance image of braincontaining a tumor. The detection is based on lo-cating the area that breaks the left-right symmetry ofthe brain. The created algorithm was tested on 73images containing tumor, tumor with edema or onlyedema. These pathological structures had varioussizes and shapes and were located in various partsof the brain.