基于血管内超声的动脉斑块识别
|
摘要
冠状动脉斑块在血管内超声图像上主要表现为内膜和内膜下组织不同程度地增厚,导致管腔横截面积缩小。识别动脉斑块的类型,可以为临床治疗提供指导意义。动脉斑块主要分为脂质斑块、纤维斑块和钙化斑块,其回声强度依次增强。利用图像像素点灰度在空间的分布规律提取动脉斑块图像纹理特征信息,选取灰度共生矩阵(GLCM)、局部二值模式(LBP)、邻域灰度(NGL)3种特征提取方式,并对提取到的特征通过支持向量机(SVM)和纠错输出码(ECOC)进行分类。结果表明,3种特征提取方式组合的分类准确性上获得较好的效果。
The atherosclerotic plaque in the intravascular ultrasound image is mainly thickened by the intima and endominal tissue, which leads to the narrowing of the transverse section of the lumen and the identification of the types of atherosclerotic plaques, which can provide guidance for clinical treatment. Atherosclerotic plaques were mainly composed of lipid plaques, fibrous plaques and calcified plaques, and their echo intensity increased in turn. The texture feature information of arterial plaque image is extracted from the distribution of image pixel gray level in space, and three kinds of feature extraction methods are selected, such as GLCM, LBP, NGL. The extracted features are classified by the support vector machine(SVM) and the error correcting output codes(ECOC). The results show that the classification accuracy of the three feature extraction methods is better.
The atherosclerotic plaque in the intravascular ultrasound image is mainly thickened by the intima and endominal tissue, which leads to the narrowing of the transverse section of the lumen and the identification of the types of atherosclerotic plaques, which can provide guidance for clinical treatment. Atherosclerotic plaques were mainly composed of lipid plaques, fibrous plaques and calcified plaques, and their echo intensity increased in turn. The texture feature information of arterial plaque image is extracted from the distribution of image pixel gray level in space, and three kinds of feature extraction methods are selected, such as GLCM, LBP, NGL. The extracted features are classified by the support vector machine(SVM) and the error correcting output codes(ECOC). The results show that the classification accuracy of the three feature extraction methods is better.
引文
[1] 周玉杰. 冠状动脉血管内超声的临床应用与实践[M].北京:人民卫生出版社,2008.
[2] 高晓飞,蒋晓敏,肖平喜,等.血管内超声在冠状动脉疾病诊断和介入治疗中的应用[J].现代生物医学进展,2015,15(24):4790-4793.
[3] 舍恩哈根,德弗朗科,尼森,等. 轻松掌握血管内超声[M]. 北京:人民军医出版社,2009.
[4] 王树鹏,王悦喜.血管内超声在冠心病介入诊断和治疗中的应用进展[J].中国循证心血管医学杂志,2016,8(10):1269-1272.
[5] STARY HC, CHANDLER AB, DINSMORE RE, et al. A report from the committee on vascular lesions of the council on arteriosclerosis, American heart association[J]. Circulation, 1995, 92(5):1355.
[6] 冯建辉,杨玉静.基于灰度共生矩阵提取纹理特征图像的研究[J].北京测绘,2007(3):19-22.
[7] HARALICK R, SHANMUGAM K, DINSTEIN I. Textural freatures of image classification[J]. IEEE Trans Syst Man Cyber, 1973(6):610-621.
[8] 黄志杰,王伊侬,王青.基于血管内超声图像的心血管动脉粥样硬化斑块组织自动定征的研究[J].计算机科学,2018,45(5):260-265.
[9] ULABY FT, KOUVATE F, BRISCO B, et al. Textural information in SAR Images[J]. IEEE Transactions on Geoscience and Remote Sensing, 1986,24(2):235-245.
[10] 宋克臣,颜云辉,陈文辉,等.局部二值模式方法研究与展望[J].自动化学报,2013,39(6):730-744.
[11] OJALA T. Multiresolution gray-scale and rotation invariant texure[J]. Classification with Local Binary Patterns, 2002, 24(7):971-987.
[12] TAKI A, HETTERRICH H, ROODAKI A. A new approach for improving coronary plaque component analysis based on intravascular ultrasound images[J]. Ultrasound Med. Biol.2010,36(8):1245-1258.
[13] FERRARI V. Libsvm: A library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology,2008.
[14] 汪友生,吴焕焕,胡百乐,等.基于灰度共生矩阵的动脉斑块图像特征分析[J].电子测量技术,2012,35(2):55-58.
[15] BAUTISTA Má, ESCALERA S, et al. Minimal design of error-correcting output codes[J]. Pattern Recognition Letters,2012,33(6):693-702.
[2] 高晓飞,蒋晓敏,肖平喜,等.血管内超声在冠状动脉疾病诊断和介入治疗中的应用[J].现代生物医学进展,2015,15(24):4790-4793.
[3] 舍恩哈根,德弗朗科,尼森,等. 轻松掌握血管内超声[M]. 北京:人民军医出版社,2009.
[4] 王树鹏,王悦喜.血管内超声在冠心病介入诊断和治疗中的应用进展[J].中国循证心血管医学杂志,2016,8(10):1269-1272.
[5] STARY HC, CHANDLER AB, DINSMORE RE, et al. A report from the committee on vascular lesions of the council on arteriosclerosis, American heart association[J]. Circulation, 1995, 92(5):1355.
[6] 冯建辉,杨玉静.基于灰度共生矩阵提取纹理特征图像的研究[J].北京测绘,2007(3):19-22.
[7] HARALICK R, SHANMUGAM K, DINSTEIN I. Textural freatures of image classification[J]. IEEE Trans Syst Man Cyber, 1973(6):610-621.
[8] 黄志杰,王伊侬,王青.基于血管内超声图像的心血管动脉粥样硬化斑块组织自动定征的研究[J].计算机科学,2018,45(5):260-265.
[9] ULABY FT, KOUVATE F, BRISCO B, et al. Textural information in SAR Images[J]. IEEE Transactions on Geoscience and Remote Sensing, 1986,24(2):235-245.
[10] 宋克臣,颜云辉,陈文辉,等.局部二值模式方法研究与展望[J].自动化学报,2013,39(6):730-744.
[11] OJALA T. Multiresolution gray-scale and rotation invariant texure[J]. Classification with Local Binary Patterns, 2002, 24(7):971-987.
[12] TAKI A, HETTERRICH H, ROODAKI A. A new approach for improving coronary plaque component analysis based on intravascular ultrasound images[J]. Ultrasound Med. Biol.2010,36(8):1245-1258.
[13] FERRARI V. Libsvm: A library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology,2008.
[14] 汪友生,吴焕焕,胡百乐,等.基于灰度共生矩阵的动脉斑块图像特征分析[J].电子测量技术,2012,35(2):55-58.
[15] BAUTISTA Má, ESCALERA S, et al. Minimal design of error-correcting output codes[J]. Pattern Recognition Letters,2012,33(6):693-702.