为克服手工判别动脉粥样硬化易损斑块耗时耗力、主观性强、重复性差等缺点，研究了基于血管内超声自动识别易损斑块的方法。首先将Contourlet变换与Snake模型相结合进行斑块的图像分割，提取内腔轮廓与外弹力膜。接着实现经典形态特征的计算机自动提取，并提取纹理、弹性两类新特征以量化斑块属性，其中纹理特征包括一阶统计量和灰度共生矩阵特征，弹性特征的提取则基于非刚性图像配准。最后设计Fisher线性判别、支撑向量机、广义相关学习矢量量化3种分类器进行分类判决。对124例斑块(36例易损，88例非易损)的实验结果表明: 20个形态特征、24个纹理特征和6个弹性特征在两类斑块间存在显著性差异(P<0.05); 采用三类特征由支撑向量机进行分类时效果最好，在测试集上敏感性、特异性、准确率和约登指数分别达到91.7%、97.7%、96.7%和89.4%，表明利用血管内超声图像中斑块的三类特征能自动、准确地识别易损斑块。

In order to overcome drawbacks in manual identification of vulnerable atherosclerotic plaques, a method for automatic identification of vulnerable plaques is proposed based on computerized analysis of intravascular ultrasound images. First, the Contourlet transform is combined with the Snake model to segment images and detect lumen borders and external elastic membranes. Two categories of new features representing texture and elasticity of plaques are then automatically extracted to quantitate the features of plaques. The texture features consist of first-order statistics and featuresfrom the gray-level coocurrence matrix, and the elastic features are extracted from strain tensors estimated by nonrigid image registration. Finally, three types of features are used to design classifiers including Fisher linear discrimination, support vector machines, and generalized relevance learning vector quantization. The experimental results on 124 plaques, consisting of 36 vulnerable and 88 nonvulnerable ones, reveals that 20 morphological features, 24 texture features and 6 elastic features has significant difference (P<0.05) between the two types of plaques.The Support Vector Machine(SVM) outperformes the other two classifiers with the sensitivity, specificity, correct rate, and Youden’s index of 91.7%, 97.7%, 96.7%, and 89.4%, respectively. Therefore, the proposed method can automatically and accurately identify vulnerable plaques.