计算机科学 ›› 2018, Vol. 45 ›› Issue (5): 260-265.doi: 10.11896/j.issn.1002-137X.2018.05.045
所属专题: 医学图像
黄志杰,王伊侬,王青
HUANG Zhi-jie, WANG Yi-nong and WANG Qing
摘要: 为了获取患者心血管内斑块特征的准确信息,并辅助临床医生对动脉粥样硬化区域进行判断和识别,文中进行了基于血管内超声(IVUS)图像的心血管粥样硬化斑块组织自动定征的研究。本研究收集了10个心血管疾病患者的IVUS图像,共207块斑块样本。首先,确定滑动邻域块的尺寸,令其中心像素遍历斑块区域,遍历过程中计算每个滑动邻域块的灰度均值和熵,并沿4个方向运用灰度共生矩阵法求出共生矩阵的10个局部特征;然后,对IVUS图像进行Gabor滤波和局部二值模式(LBP)处理,获得了更多的图像纹理特征;最后,通过线性分类器Liblinear、随机森林分类器(Random Forests)和调和最小值-广义学习向量量化分类器(H2M-GLVQ)对降维后的特征数据进行分类判决。将医生人工标记的结果作为金标准,自动定征的实验结果表明,随机森林和H2M-GLVQ分类器总体上对斑块组织的识别准确率均达到80%以上,其中随机森林分类器识别纤维化、脂质和钙化样本斑块的平均识别准确率分别为89.04%,80.23%和73.77%。
[1] HANSSON G K.Inflammation,atherosclerosis,and coronaryartery disease[J].The New England Journal of Medicine,2005,352(16):1685-1695. [2] VIRMANI R,KOLODGIE F D,BURKE A P,et al.Atherosclerotic plaque progression and vulnerability to rupture-angiogenesis as a source of intraplaque hemorrhage[J].Arteriosclerosis Thrombosis and Vascular Biology,2005,25(10):2054-2061. [3] JEERS S,NARBUTE I,ERGLIS A.Use of intravascular imaging in managing coronary artery disease[J].World Journal of Cardiology,2014,6(6):393-404. [4] NAIR A,KUBAN B D,OBUCHOWSKI N,et al.Assessingspectral algorithms to predict atherosclerotic plaque composition with normalized and raw intravascular ultrasound data[J].Ultrasound in Medicine Biology,2001,27(10):1319-1331. [5] KATOUZIAN A,SATHYANARAYANAN S,BASERI B,et al.Challenges in atherosclerotic plaque characterization with intravascular ultrasound:From data collection to classification[J].IEEE Transactions on Information Technology in Biomedicine:A Publication of IEEE Engineering in Medicine and Biology Society,2008,12(3):315-327. [6] NAIR A,KUBAN B D,TUZCU E M,et al.Coronary plaque classification with intravascular ultrasound radiofrequency data analysis[J].Circulation,2002,106(17):2200-2206. [7] MAURICE R L,FROMAGEAU J,BRUSSEAU L,et al.On the potential of the lagrangian estimator for endovascular ultrasound elastography:In vivo human coronary artery study[J].Ultrasound in Medicine and Biology,2007,33(8):1199-1205. [8] ARAKI T,IKEDA N,SHUKLA D,et al.PCA-based pollingstrategy in machine learning framework for coronary artery di-sease risk assessment in intravascular ultrasound:A link between carotid and coronary grayscale plaque morphology[J].Computer Methods and Programs in Biomedicine,2016,128:137-158. [9] LO VERCIO L,ORLANDO J I,DEL FRESNO M,et al.Assessment of image features for vessel wall segmentation in intravascular ultrasound images[J].International Journal of Computer Assisted Radiology and Surgery,2016,11(8):1397-1407. [10] ATHANASIOU L S,KARVELIS P S,TSAKANIKAS V D,et al.A novel semiautomated atherosclerotic plaque characteri-zation method using grayscale intravascular ultrasound images:Comparison with virtual histology[J].IEEE Transactions on Information Technology in Biomedicine,2012,16(3):391-400. [11] JABASON E.Performance analysis of contourlet features with SVM classifier for the characterization of atheromatous plaque in intravascular ultrasound images[J].International Journal of Engineering Research & Applications,2014,4(3):35-42. [12] ZHANG J,TAN T,MA L.Invariant texture segmentation via circular gabor filters[C]∥IEEE International Conference on Pattern Recognition.2002:901-904. [13] PIETIK O T,INEN M,et al.Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2002,24(7):971-987. [14] GIANNOGLOU V G,STAVRAKOUDIS D G,T HEOCHAIRS J B.IVUS-based characterization of atherosclerotic plaques using feature selection and SVM classification[C]∥IEEE International Conference on Bioinformatics & Bioengineering.2012:715-720. [15] FAN R E,CHANG K W,HSIEH C J,et al.Liblinear:A library for large linear classification[J].Journal of Machine Learning Research,2008,9(9):1871-1874. [16] BREIMAN L.Random forests[J].Machine Learning,2001,45(1):5-32. [17] QIN A K,SUGANTHAN P N.Rapid and brief communication:Initialization insensitive LVQ algorithm based on cost-function adaptation[J].Pattern Recognition,2005,38(5):773-776. [18] Sato A.Generalized learning vector quantization[C]∥Confe-rence on Neural Information Processing Systems.1996:423-429. [19] SUN Z,WANG L X,ZHOU Y.Automated tissue characterization of intravascular ultrasound gray-scale images[J].Journal of Biomedical Engineering,2016,33(2):287-294.(in Chinese) 孙正,王立欣,周雅.血管内超声灰阶图像的自动组织标定[J].生物医学工程学杂志,2016,33(2):287-294. [20] HARALICK R M.Texture features for image classification[J].IEEE Transactions on Systems Man & Cybernetics,1990,3(6):610-621. [21] SCHOENHAGEN P,CROWE T,NICHOLL S,et al.IVUSmake easy[M].America,Paul G.Informa Healthcare,2008. [22] QIN A K,SUGANTHAN P N,LIANG J J.A new generalized LVQ algorithm via harmonic to minimum distance measure transition[C]∥IEEE International Conference on Systems,Man &Cybernetics.2004:4821-4825. |
No related articles found! |
|