全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

采用加性核SVM的二尖瓣瓣根识别

DOI: 10.11834/jig.20140509

Keywords: 二尖瓣,上下文特征,K-Means,SVM分类器,加性核

Full-Text   Cite this paper   Add to My Lib

Abstract:

目的超声心动图中图像噪声严重、分辨率低以及成像范围有限等缺点,导致二尖瓣(MA)瓣根的识别非常困难,采用加性核函数的支持向量机(SVM)分类器识别超声心动图中的二尖瓣瓣根位置。方法心脏二尖瓣瓣根位置对于心室的分割、心脏建模以及多模态配准很重要。本文提出将加性核支撑向量机分类算法并结合一个局部的上下文特征用于二尖瓣瓣根的识别。主要创新点有:1)利用图像中的上下文特征提取二尖瓣瓣根部特征;2)应用最小加性核的SVM分类器快速识别二尖瓣瓣根的候选点;3)对于候选点应用加权模板,计算候选点的加权密度;4)在加权密度场中,采用二分查找算法,自适应确定一个阈值,剔除二尖瓣瓣根的错分点,确定二尖瓣瓣根的位置。结果本文算法在10个儿科病人的超声四腔心动图上测试,和手动选出的二尖瓣瓣根点相比,平均误差控制在1.52±2.25个像素。结论采用加性核函数的SVM分类器能够快速、准确地识别二尖瓣瓣根点。

References

[1]  Eto Y,Yamada H,Shin J H,et al. Automated mitral annular tracking: a novel method for evaluating mitral annular motion using two dimensional echocardiography[J]. AM Soc Echocardiogr, 2005, 18(4):306-312.
[2]  Pai R G,Bodenheimer M M,Pai S M,et al. Usefulness of systolic excursion of the mitral annulus as an index of left ventricular systolic function[J]. The American Journal of Cardiology, 1991, 67(2):222-224.
[3]  Willenheimer R,Israelsson B,Cline C,et al. Left atrioventricular plane displacement is related to both systolic and diastolic left ventricular performance in patients with chronic heart failure[J]. Eur. Heart J., 1999,20(8):612- 618.
[4]  Nevo S T,Van Stralen M,Vossepoel A M,et al. Automated tracking of the mitral valve annulus motion in apical echocardiographic images using multidimensional dynamic programming[J]. Ultrasound Med, Biol., 2007,33(9):1389-1399.
[5]  Takemoto Y,Hozumi T,Sugioka K,et al. Automated three-dimensional analysis of mitral annular dynamics in patients with myocardial infarction using automated mitral annular tracking method[J]. Echocardiography, 2006,23(8):658-665.
[6]  Veronesi F,Corsi C,Caiani E,et al. Semi-automatic tracking for mitral annulus dynamic analysis using real-time 3D echocardiography[J]. Computers in Cardiology, 2006,33:113-116.
[7]  Schneidera R J,Perrina D P, Vasilyevb N V, et al. Mitral annulus segmentation from four-dimensional ultrasound using a valve state predictor and constrained optical flow[J]. Medical Image Analysis, 2012,16(2):497-504.
[8]  Schneider R,Perrin D,Vasilyev N,et al. Mitral annulus segmentation from 3D ultrasound using graph cuts[J]. IEEE Trans. on Med. Imaging, 2010,29(9):1676-1687.
[9]  Maji S,Berg A C,Malik J. Efficient classification for additive kernel SVMs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(1):66-77.
[10]  Maji S,Berg A C,Malik J. Classification using intersection kernel support vector machines is efficient[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, AK: IEEE,2008:1-8.
[11]  He D C,Wang L. Texture unit, texture spectrum,and texture analysis,geoscience and remote sensing[J]. IEEE Transactions Geoscience and Remote Sensing, 1990,28(4):509-512.
[12]  Wang L,He D C. Texture classification using texture spectrum[J]. Pattern Recognition,1990,23(8):905-910.
[13]  Heisele B,Serre T,Prentice S,et al. Hierarchical classification and feature reduction for fast face detection with support vector machines[J]. Pattern Recognition,2003,36(9):2007-2017.
[14]  Tu Z W,Bai X. Auto-Context and its application to high-level vision tasks and 3D brain image segmention[J]. IEEE Transactions on pattern Analysis and Machine Intelligence, 2010,32(10):1744-1757.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133