基于K-最近邻规则的磁共振颅脑图像分割算法的应用研究

Application Study of Segmentation Algorithm of Head MRI Based ON K-Nearest Neighbor Rule

目的 介绍一种简单实用的磁共振颅脑图像分割算法K 最近邻 (简称K NN)规则 ,并利用该算法对磁共振颅脑图像进行分割研究。方法 该方法是一个多步处理过程。首先利用边界跟踪法对磁共振颅脑图像进行预处理 ,剔除颅骨和肌肉等非脑组织 ,只保留大脑结构 ;然后利用K NN规则对大脑结构进行分割 ,从大脑结构中分别提取出白质 (WM)、灰质 (GM)和脑脊液 (CSF)。结果 分割算法在预处理步中能精确地分割出大脑结构 ,在K NN分割步中能很好地从大脑结构中分割出WM、GM和CSF。结论 该算法在磁共振颅脑图像的分割中简单实用 ,具有很强的鲁棒性和稳定性。

Purpose A simple and practical head MRI segmentation algorithm, K nearest neighbor ( K? NN, for short) rule, was introduced in this paper. Methods It is a procedure of multi-step processing. At first, the non-brain tissues, such as skull and muscle, were moved by edge-tracing method, and the brain structure can be remained; and then, the white matter(WM), gray matter(GM)and cerebral spinal fluid(CSF)can be separated from the brain structure by using K-NN rule to segment the brain structure. Results The segmentation method put forth by this paper can correctly separate brain structure in the ima ge preprocessing step, and then the WM, GM and CSF can be segmented better by the K-NN rule. Conclusions The K-NN rule is simple and practical and has strong robustness in the segmentation of MRI.