Unsupervised Abnormalities Extraction and Brain Segmentation

摘要：

In this paper,we propose a methodology consists ofseveral unsupervised clustering techniques to acquirea satisfactory segmentation of Computed Tomography(CT)brain images.The ultimate goal of segmentationis to obtain three segmented images,which are theabnormalities,cerebrospinal fluid(CSF)and brainmatter respectively.The proposed approach containsof two phase-segmentation methods.In the first phasesegmentation,the combination of k-means and fuzzy c-means(FCM)methods is implemented to partition theimages into the binary images.From the binaryimages,a decision tree is then utilized to annotate theconnected component into normal and abnormalregions.For the second phase segmentation,theobtained experimental results have shown thatmodified FCM with population-diameterindependent(PDl)segmentation is more feasible andyield satisfactory results.

作者: Tong Hau Lee Mohammad Faizal Ahmad Fauzi Ryoichi Komiya Su-Cheng Haw

作者单位: Faculty of Information Technology Multimedia University,Jalan Multimedia,63100 Cyberjaya,Selangor,Ma Faculty of Engineering,Multimedia University,Jalan Multimedia,63100 Cyberjaya,Selangor,Malaysia.

会议类型: 国际会议

会议名称: 2008 3rd International Conference on Intelligent System and Knowledge Engineering(第三届智能系统与知识工程国际会议)(ISKE 2008)

会议地点: 厦门

会议语种:英文

页码: 1185-1190

在线出版日期: 2008-11-17（万方平台首次上网日期，不代表论文的发表时间）