Image Mining Automata Based Seeded Tumor C-Taxonomy Algorithm for Segmentation of Brain Tumors on MR Images (BITA)Author : P.Senthil
Volume 5 No.1 January-June 2016 pp 10-16
In this paper, CA algorithm is used to establish the connection of the CA-based segmentation to the graphtheoretic methods to show that the iterative CA framework solves the shortest path problem with proper choice of transition rule. An algorithm based on CA is
presented to differentiate necrotic and enhancing tumor tissue content to assist clinicians and researchers in radiosurgery planning and assessment of the response to the therapy. Proposed segmentation framework is composed of three stages. First VOI is selected
with foreground & background seeds using the line drawn by the user over the largest visible diameter of the tumor. In second stage, tumor CA algorithm is run on the VOI for the foreground & background seeds to obtain strength maps. Two strength maps are combined to obtain tumor probability map & level set surface is evolved on tumor probability map to impose spatial smoothness. Finally necrotic regions of the tumor is segmented using CA based method with chosen enhanced & necrotic seeds.
Tumor segmentat ion, Cellular Aut omat a (CA), Magnetic Resonance Imaging (MRI), Necrotic region, Radiotherapy, Seeded segmentation.