一种磁共振脑组织及肿瘤区域的提取方法

摘要

The brain tissue extraction in brain magnetic resonance images ( MRI) improves the accuracy of the brain region of interest, and the image segmentation of brain tumors can be used for tissue reconstruction, tumor volume calculation, etc. In this paper, we first used the brain extraction tool( BET) method to extract brain structure region, then combined the automatic entropy threshold and morphology to extract. After that , we applied the fuzzy C-means ( FCM ) algorithm and median filter to extract the tumor region. Compared with the result of expert sketch, this paper preferable extracts brain tumor region, actualizes the automatic segmentation, and improves the calculating efficiency.%脑成像的脑组织准确提取有助于提高研究脑部感兴趣区域的精度,脑部肿瘤图像分割的研究可用于组织三维重建、肿瘤体积计算等.本文使用磁共振图像,首先采用脑提取(brain extraction tool,BET)算法提取出脑组织区域,再利用一种信息熵自动阈值与形态学结合的方法;当提取出脑组织后,利用模糊C均值(fuzzy C-means,FCM)聚类算法对肿瘤区域进行提取,经过中值滤波等步骤,提取出脑肿瘤区域.与专家手工勾画对比,本文较好地提取出了脑肿瘤组织,从而实现了脑组织的自动分割,提高了工作效率.