基于高斯混合分布和区域竞争主动轮廓的医学目标分割

提出了一种基于高斯混合分布和区域竞争主动轮廓的医学目标提取模型。这一模型,把主动轮廓的能量函数表示为像素属于目标或背景的子类的最大概率的区域积分,在水平集合框架下使能量函数最小化,得到在高斯子类区域间竞争演化的分割迭代方程。同时,附加的速度约束项使主动轮廓在越过目标边缘时速度降低,提高了分割的收敛性。通过大量肝脏CT图像的分割实验以及与几种经典模型和手工提取的比较,表明该模型在医学图像分割中具有较好的健壮性、准确性和灵活性均较好。

Medical Object Extraction by Gaussian Mixture Model and Region Competition Active Contour Model

This paper proposed a regional active contour model with an embedded classifier, based on a Gaussian mixlure model fitted to the intensity distribution of the medical image. The difference between the maximum probability of the intensities belonging to the classes or subclasses of the object and those of the background is made as an energy term in the active contour modcl,and minimization of the whole energy function leads to a novel iterative equation. An additional speed controlling term slows down the evolution of the active contour when it approaches an edge, making it quickly convergent to the ideal object. The developed model has been applied to liver segmentation. Some comparisons are made between the geodesic active contour,C-V(active contour without edges),manual outline and our model. As the experiments show that our model is accurate,flexible and suited to extract objects surrounded by a complicated back-ground.