Brain Tumor Segmentation Using OTSU Embedded Adaptive Particle Swarm Optimization Method and Convolutional Neural Network

Medical imaging and deep learning have tremendously shown improvement in research field of brain tumor segmentation. Data visualization and exploration plays important role in developing robust computer aided diagnosis system. The analysis is performed in proposed work to provide automation in brain tumor segmentation. The adaptive particle swarm optimization along with OTSU is contributed to determine the optimal threshold value. Anisotropic diffusion filtering is applied on brain MRI to remove the noise and improving the image quality. The extracted features provided as data for training the convolutional neural network and performing classification. The proposed research achieves higher accuracy of 98% which is better over existing methods.