曲阜师范大学学报（自然科学版）2018，Vol.25Issue(2)：49-53,5.DOI:10.3969/j.issn.1001-5337.2018.2.049
基于卷积神经网络特征提取的MRI脑肿瘤图像分割
Brain Tumor Segmentation Using Convolutional Neural Networks Feature Extraction in MRI Images
摘要
Abstract
Brain tumor segmentation plays an important role in assisting diagnosis,planning treatment and surgical navigation.In this paper,a segmentation method based on convolution neural network feature extraction to segment the lesion of the tumor is proposed.Our algorithm is composed of two cascaded sta-ges.In the first stage,we train CNN to learn the mapping from the image space to the tumor label space. During the testing phase,we use the predicted label output from CNN and send it along with the testing image to an SVM classifier for accurate segmentation.The experimental results show that the method can adapt to the difference of brain tumor,and the maximum segmentation accuracy is up to 9 3 %.关键词
卷积神经网络/支持向量机/特征提取/脑肿瘤分割Key words
convolutional neural networks/support vector machine/feature extraction/brain tumor segmentation分类
信息技术与安全科学引用本文复制引用
谢铭超,韩旭,栾帅,李芳,王春兴..基于卷积神经网络特征提取的MRI脑肿瘤图像分割[J].曲阜师范大学学报（自然科学版）,2018,25(2):49-53,5.基金项目
国家自然科学基金(11674198). （11674198）
