Brain Tumor Segmentation based on 3D Neighborhood Features Using Rule-based Learning

摘要

In order to plan precise treatment or accurate tumor removal surgery, brain tumor segmentation is critical fordetecting all parts of tumor and its surrounding tissues. To visualize brain anatomy and detect its abnormalities, we usemulti-modal Magnetic Resonance Imaging (MRI) as input. This paper introduces an efficient and automated algorithmbased on the 3D bit-plane neighborhood concept for Brain Tumor segmentation using a rule-based learning algorithm. Inthe proposed approach, in addition to using intensity values in each slice, we consider sets of three consecutive slices toextract information from 3D neighborhood. We construct a Rule base using sequential covering algorithm. Through arule-based ordering method and a reward/penalty policy, we assign weights to each rule such that the largest weight isassigned to the strongest (mostly referred) rule. Finally, the rules are ranked from the strongest to the weakest. Regardingto the strength of rules in the framework, those with highest weight are selected for voxel labeling. This algorithm istested on BRATS 2015 training database of High and Low Grade tumors. Dice and Jaccard indices are calculated andcomparative analysis is implemented as well. Experimental results indicate competitive performance compared to thestate of the art methods.