基于一类支持向量机的冠脉病变检测方法

针对冠脉病变检测算法普遍存在的异常截面识别率低、无法排除特殊结构影响等问题，提出了一种基于一类支持向量机(OCSVM)的冠脉病变检测方法，并使用冠脉面重采样和基于最大互信息的特征选择方法提高了算法识别正确率。该方法首先基于梯度通量对冠脉源截面进行三次样条插值重采样，然后构造出截面的多尺度特征，接着使用最大互信息结合冗余度去除进行特征选择，最后使用特征数据训练OCSVM完成冠脉病变检测。实验结果显示，在1128个冠脉截面数据的测试结果中，本算法在完全识别异常截面的情况下对健康截面的识别正确率达到了53.5%，远高于同类型的仅从正面和未标记数据学习的支持向量机(SVM)算法所对应的19.6%；而冠脉截面重采样也使得30个特征数下算法对健康截面的识别正确率由21.7%提高到了53.2%。

Abstract

To solve problems such as low recognition rate of abnormal cross section and inability to rule out special structural effects, a method based on one-class support vector machine (OCSVM) is proposed to detect coronary lesion. By using coronary cross section resampling and feature selection based on maximum mutual information, the method achieves a relatively high recognition rate. At first, the coronary cross section is resampled based on gradient flux using cubic spline interpolation, and multi-scale feature vector is constructed for every coronary cross section. Then, a maximum mutual information method combined with redundancy removal is adopted to select target features. Finally, selected features are used to train OCSVM model to complete coronary lesion detection. The experiment results in 1128 cross section data show that the maximal recognition rate of the proposed method of health cross section reaches 53.5%, which is much higher than that of support vector machine (SVM) algorithm (learning only from positive and unlabeled data) of 19.6%, with complete recognition of abnormal cross section. Meanwhile, the health cross section recognition rate by 30 features rises from 21.7% to 53.2% owing to the resampling of the cross section.

参考文献

[1] 陈伟伟, 高润霖, 刘力生, 等. 《中国心血管病报告2015》概要［J］. 中国循环杂志, 2016, 31(6): 617-622.

Chen Weiwei, Gao Runlin, Liu Lisheng, et al. Outline of 《China cardiovascular disease report 2015》［J］. Chinese Circulation Journal, 2016, 31(6): 617-622.

[2] Maurovich-Horvat P, Ferencik M, Voros S, et al. Comprehensive plaque assessment by coronary CT angiography［J］. Nature Reviews Cardiology, 2014, 11(7): 390-402.

[3] 陈相廷, 张帆, 张一凡, 等. CT造影冠状动脉狭窄检测与量化的相关研究［J］. 激光与光电子学进展, 2015, 52(8): 080006.

Chen Xiangting, Zhang Fan, Zhang Yifan, et al. Research on stenosis detection and quantification of coronary artery in CT angiography［J］. Laser & Optoelectronics Progress, 2015, 52(8): 080006.

[4] 李磊, 韩玉, 席晓琦, 等. X射线CT转台单侧两次螺旋扫描单层重排重建算法［J］. 光学学报, 2016, 36(9): 0934001.

Li Lei, Han Yu, Xi Xiaoqi, et al. Single-slice-rebinning reconstruction algorithm with one-sided two helical scans for X-ray CT system［J］. Acta Optica Sinica, 2016, 36(9): 0934001.

[5] Kang D, Slomka P J, Nakazato R, et al. Automated knowledge-based detection of nonobstructive and obstructive arterial lesions from coronary CT angiography［J］. Medical Physics, 2013, 40(4): 041912.

[6] Tessmann M, Vega-Higuera F, Fritz D, et al. Multi-scale feature extraction for learning-based classification of coronary artery stenosis［C］. SPIE, 2009, 7260: 726002.

[7] Freund Y, Schapire R E. A decision-theoretic generalization of on-line learning and an application to boosting［J］. Journal of Computer & System Sciences, 1997, 55(1): 119-139.

[8] Zuluaga M A, Magnin I E, Hernández H M, et al. Automatic detection of abnormal vascular cross-sections based on density level detection and support vector machines［J］. International Journal of Computer Assisted Radiology and Surgery, 2011, 6(2): 163-74.

[9] Zuluaga M A, Hoyos M H, Orkisz M. Feature selection based on empirical-risk function to detect lesions in vascular computed tomography［J］. IRBM, 2014, 35(5): 244-254.

[10] Zuluaga M A, Hush D, Leyton E J F D, et al. Learning from only positive and unlabeled data to detect lesions in vascular CT images［C］. International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer-Verlag, 2011: 9-16.

[11] 李俊威, 陈晓冬, 贾忠伟, 等. 基于光线投射法的冠脉血管腔壁分割算法［J］. 中国激光, 2015, 42(8): 0804002.

Li Junwei, Chen Xiaodong, Jia Zhongwei, et al. A coronary artery lumen segmentation algorithm based on ray casting［J］. Chinese J Lasers, 2015, 42(8): 0804002.

[12] 张朝霞, 陈晓冬, 单建丰, 等. 基于多层螺旋CT血管分析的感兴趣冠脉段最佳造影角度计算［J］. 中国激光, 2011, 38(11): 1104003.

Zhang Zhaoxia, Chen Xiaodong, Shan Jianfeng, et al. Calculation of optical angiographic angle for segment of interest based on multislice computed tomography vessel analysis［J］. Chinese J Lasers, 2011, 38(11): 1104003.

[13] Lesage D,Angelini E D, Bloch I, et al. A review of 3D vessel lumen segmentation techniques: models, features and extraction schemes［J］. Medical Image Analysis, 2009, 13(6): 819-845.

[14] Zuluaga M A, Leyton E J F D, Hoyos M H, et al. Feature selection for SVM-based vascular anomaly detection［C］. Medical Computer Vision. Recognition Techniques and Applications in Medical Imaging - International Miccai Workshop, Mcv. 2010: 141-152.

[15] 辛竹, 周亚建. 文本分类中互信息特征选择方法的研究与算法改进［J］. 计算机应用, 2013, 33(s2): 116-118.

Xin Zhu, Zhou Yajian. Study and improvement of mutual information for feature selection in text categorization［J］. Journal of Computer Applications, 2013, 33(s2): 116-118.

[16] 姚旭, 王晓丹, 张玉玺, 等. 特征选择方法综述［J］. 控制与决策, 2012, 27(2): 161-166.

Yao Xu, Wang Xiaodan, Zhang Yuxi, et al. Summary of feature selection algorithms［J］. Control and Decision, 2012, 27(2): 161-166.

[17] Battiti R. Using mutual information for selecting features in supervised neural net learning［J］. IEEE Transactions on Neural Networks, 1994, 5(4): 537-550.

[18] 王皓, 孙宏斌, 张伯明. PG-HMI: 一种基于互信息的特征选择方法［J］. 模式识别与人工智能, 2007, 20(1): 55-63.

Wang Hao, Sun Hongbin, Zhang Boming. PG-HMI: mutual information based feature selection method［J］. Pattern Recognition and Artificial Intelligence, 2007, 20(1): 55-63.

[19] Schlkopf B, Smola A, Williamson R, et al. New support vector algorithms［J］. Neural Computation, 2000, 12(5): 1207-1245.

[20] Cortes C, Vapnik V. Support vector network［J］. Machine Learning, 1995, 20(3): 273-297.

[21] Avila L S, Barre S, Blue R, et al. The VTK user′s guide［M］. Clifton Park: Kitware, 2010.

[22] Chen Y W, Lin C J. Combining SVMs with various feature selection strategies［J］. Studies in Fuzziness & Soft Computing, 2006, 207: 315-324.

赵聪, 陈晓冬, 张佳琛, 汪毅, 贾忠伟, 陈向志, 郁道银. 基于一类支持向量机的冠脉病变检测方法[J]. 中国激光, 2017, 44(5): 0504006. Zhao Cong, Chen Xiaodong, Zhang Jiachen, Wang Yi, Jia Zhongwei, Chen Xiangzhi, Yu Daoyin. Coronary Lesion Detection Method Based on One-Class Support Vector Machine[J]. Chinese Journal of Lasers, 2017, 44(5): 0504006.

基于一类支持向量机的冠脉病变检测方法

关于本站 Cookie 的使用提示

全站搜索

基于一类支持向量机的冠脉病变检测方法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索