Segmenting MRI Brain Tumor Images Using Modified Cuckoo Search Optimization with Morphological Reconstruction Filters

Cancer is dangerous diseases and it affects the entire body if necessary steps are taken to prevent the starting stage of tumor, to measure the tumor growth and ensure that tumor is malignant or benign is a challenging task for the radiologist. In our work modified cuckoo search algorithm with morphological reconstruction filters is used to segment the MR brain tumor. All images are collected BRATS 2013 and clinical dataset for proving the efficiency of the algorithm. Segmentation performance measures like Entropy, Mean Square Error, Peak Signal to Noise Ratio, Precision, Recall, and accuracy values are calculated to ensure the accuracy of the proposed work. Our work yields 97 % accuracy which is far better than the existing techniques like Particle Swarm optimization, Genetic Algorithm and Ant Colony Optimization.