基于马尔可夫场的多发性硬化症MR图像分割算法

Segmentation of Multiple Sclerosis Lesions Based on Markov Random Fields Model for MR Images

多发性硬化症是一种严重威胁中枢神经功能的疾病,对其病灶的分割方法研究正受到越来越多的关注。本文提出了一种基于马尔可夫场并利用多发性硬化症形态学特性的分割算法。首先运用基于MRF模型的分割算法和区域增长法,分割出脑白质所包围的区域;然后对脑白质所包围的区域再次分割,就实现了对T2加权MR脑部图像的多发性硬化症病灶分割。通过对多发性硬化症模拟和临床T2加权MR脑部图像的分割实验,表明该算法能够比较准确地分割多发性硬化症病灶,并且具有无监督、稳健性好等优点,能够应用于多发性硬化症的临床辅助诊断。

Multiple sclerosis （MS） is an inflammatory demyelinating disease that would damage central nervous system. There is a growing attention to the segmentation algorithms of MS Lesions. An MRF-based algorithm for MS lesions segmentation of T2-weighted MR brain images is developed by utilizing the morphological characteristics of MS lesion tissues. The regions circumscribed by white matter are extracted at first by MRF-based segmentation and region growing methods; the abstracted regions are then segmented again using MRF-based algorithm. The segmented MS lesions of both simulated and clinical T2-weighted MR brain images are presented in the current work. The testing results show that the proposed algorithm is robust and accurate enough for clinical use.