中文摘要：

心脏核磁共振图像分割一直是医学影像分析领域的研究热点和难点，文中提出了一种基于梯度矢量流Snake模型的左心室分割方法．作为对梯度矢量流（GVF）的改进，提出了退化最小曲面梯度矢量流（dmsGVF）．该模型对弱边界泄漏有更好的鲁棒性；挖掘了左心室的形状特点，采用相应的形状约束，克服了由于图像灰度不均而导致的局部极小，也大大减弱了分割结果对初始轮廓的依赖；对于左室壁外膜的分割，挖掘了左室壁内、外膜的位置关系，通过重新组合梯度分量来构造新的外力场．这种外力场能够克服原始梯度矢量流的不足，使得室壁外膜边缘很弱时也能得到保持，以左室壁内膜分割结果作为初始化能够自动地分割出左室壁外膜．实验结果表明，该方法能高效准确地同时分割左室壁内、外膜．

英文摘要：

Segmentation of cardiac magnetic resonance images is a hot topic in the community of medical images analysis and remains one of the open problems. In this study, the authors address this issue based on gradient vector flow snake model and make three contributions, firstly, the degenerated minimal surface GVF is proposed as an improvement on gradient vector flow, this new flow outperforms the original one concerning weak boundary leaking; secondly, the shape of the left ventricle is taken into account and a shape based energy for the snake model is adopted, with this energy, the snake contour can conquer the unexpected local minimum stemming from image inhomogeneity and the final results could depend much less on the initial contour. In order to segment the epicardium, the gradient vector components are reconfigured to generate the external force field. This new external force can overcome the demerits of the original GVF force and maintain the epicardium boundaries even if the contrast between the myocardium and neighbor organs is very low, taking the endocardium contour as initialization, the snake is reactivated to locate the epicardium accurately. The proposed strategy is validated on a large amount of cardiac magnetic resonance images.