SEM混合模型脑血管分割算法

Segmentation Algorithm of Brain Vessel Image Based on SEM Statistical Mixture Model

摘要: 针对脑血管结构复杂、空间比例小,易导致对其进行图像分割较困难的问题,面向脑MRI提出参数统计分类算法,通过随机迭代提高血管分割精度.首先应用最大强度投影法(MIP)实现脑血管图像预处理,以降低脑血管图像中混合成分的数目;其次用有限高斯混合模型模拟脑血管和脑组织的随机分布;最后通过随机期望最大化(SEM)算法进行混合模型的参数估计,解决了传统期望最大化(EM)算法收敛速度缓慢和局部极值的问题.实验结果表明,采用文中算法可有效地分割脑血管主分支及周围较细小分支,且其收敛速度比传统EM算法有较大提高.

Abstract: Due to the complicated structure and small proportion of brain vessels,a statistical analysis technology is proposed for its segmentation,and the accuracy of segmentation is improved by the random assortment iteration.First the MIP algorithm is applied to decrease the quantity of mixing elements.Then the Gauss Mixture Model is put forward to fit the stochastic distribution of the brain vessels and brain tissue.At last,the Stochastic Estimation Maximization(SEM)algorithm is adopted to estimate the parameters of Gauss Mixture Model.With the model,the small branches of the brain vessel can be segmented,the speed of the convergent is improved and local minima are avoided.The feasibility and validity of the model are verified by the experiment.