Abstract: An improved ASM method for segmentation of left ventricle tagged MR images is presented.Based on the idea of feature fusion，the Canonical Correlation Analysis（CCA） is used to combine the features extracted form tagged MR images by LM filter bank.Then，a classifier is constructed to determine edge point using SVM.CCA can decrease the classification error and improve the classification performance.Instead of sampling the normalized derivative profiles，the feature at each position along the profile perpendicular to the object contour is fed into a trained classifier to determine the edge point.The method is validated on different frame and slice of tagged MR images.Experimental results show that the method can achieve a high accurate and robust performance.

Key words: ASM model, LM filter bank, left ventricle, Tagged MR images, Canonical Correlation Analysis（CCA）

摘要： 提出了一种基于ASM框架的Tagged MR图像左心室分割方法。即从基于典型相关分析的特征融合角度对LM滤波器组提取的Tagged MR图像左心室纹理特征用典型相关分析进行优化组合，再用SVM构造分类器，通过分类器来确定边缘点，驱动ASM模型边界变形得到分割结果。通过典型相关分析的特征融合可以降低分类错误率，提高分类性能；用分类器代替经典ASM模型的基于轮廓灰度的匹配法来确定边缘点具有较强的鲁棒性。该方法在不同时刻不同断层Tagged MR图像上进行了验证，实验结果表明该方法具有较高的准确度和较强的鲁棒性。

关键词: ASM模型, LM滤波器, 左心室, Tagged MR图像, 典型相关分析