A Deep Convolutional Neural Network Learning Transfer to SVM-Based Segmentation Method for Brain Tumor

Abstract

Brain tumor segmentation plays an important role in assisting diagnosis, planning treatment and surgical navigation. In this paper, we propose a convolutional neural network-based learning transfer to support vector machine classifier method for brain tumor segmentation. Our algorithm is composed of two cascaded stages. In the first stage, we trained CNN to learn the mapping from the image space to the tumor label space. During the testing phase, we used the predicted label output from CNN and sent it along with the testing image to an SVM classifier for accurate segmentation. Then we iterate our deep CNN-SVM classifier. Experiments and comparisons demonstrate that the proposed framework is better than the separate SVM-based segmentation or CNN-based segmentation.