一种基于k-means聚类和半监督学习的医学图像分割算法

被引:5
作者
黄伟
陶俊才
机构
[1] 南昌大学信息工程学院
关键词
聚类; 相似度函数; 半监督学习; 图像分割;
D O I
10.13764/j.cnki.ncdl.2014.01.007
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
医学图像分割是计算机视觉和图像处理领域近年来研究的热点问题之一。一种基于k-means聚类和半监督学习的医学图像分割新算法被提出。在k-means聚类模型中,相似度函数是关系到聚类效果好坏的关键因素。所使用的相似度函数通过基于side-information的半监督学习方法来确定;确定后的相似度函数又被运用回k-means聚类模型中来实现对医学图像的分割。为了检验该算法效果,脑部肿瘤患者的磁共振图像被运用在实验中。分析结果表明:该算法在本文所采用的实例中能获得优于传统算法的分割效果。
引用
收藏
页码:31 / 35
页数:5
相关论文
共 7 条
[1]  
Mathematical Statistics and Data Analysis. RICE J. . 2006
[2]  
Distance Metric Learning:A Comprehensive Review. LIU Y. Miscellaneous Papers in Computer and Information Science of Michigan State University . 2006
[3]  
Distance metric learning with application to clustering with side-information. E.P.Xing,A.Y.Ng,M.I.Jordan,S.Russell. Advances in Neural Information Processing Systems 15 . 2003
[4]  
Normalized cuts and image segmentation. Shi, Jianbo,Malik, Jitendra. IEEE Transactions on Pattern Analysis and Machine Intelligence . 2000
[5]  
Pattern Classification. Duda RO,Hart PE,Stork DG. . 2001
[6]  
Semi-supervised nasopharyngeal carcinoma lesion extraction from magnetic resonance images using online spectral clustering with a learned metric. Huang Wei,Chan Kap Luk,Gao Yan,Zhou Jiayin,Chong Vincent. Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention . 2008
[7]   Region-Based Nasopharyngeal Carcinoma Lesion Segmentation from MRI Using Clustering- and Classification-Based Methods with Learning [J].
Huang, Wei ;
Chan, Kap Luk ;
Zhou, Jiayin .
JOURNAL OF DIGITAL IMAGING, 2013, 26 (03) :472-482