文献类型：会议论文

英文题名：METRIC LEARNING FOR LABEL FUSION IN MULTI-ATLAS BASED IMAGE SEGMENTATION

作者：Zhu, Hancan[1];Cheng, Hewei[2];Yang, Xuesong[3];Fan, Yong[4]

机构：[1]Shaoxing Univ, Coll Math Phys & Informat, Shaoxing 312000, Peoples R China;[2]Chongqing Univ Posts & Telecommun, Sch Bioinformat, Chongqing 400065, Peoples R China;[3]Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing, Peoples R China;[4]Univ Penn, Perelman Sch Med, Dept Radiol, Philadelphia, PA 19104 USA

会议论文集：13th IEEE International Symposium on Biomedical Imaging (ISBI)

会议日期：APR 13-16, 2016

会议地点：Prague, CZECH REPUBLIC

语种：英文

外文关键词：multi-atlas based image segmentation; hippocampus; metric learning; weighted voting; label fusion

外文摘要：A novel metric learning method is proposed to fuse segmentation labels in multi-atlas based image segmentation. Different from current label fusion methods that typically adopt a predefined distance metric model to compute the similarity between image patches of atlas images and the image to be segmented, we learn a distance metric model from the atlases to keep image patches of the same structure close to each other while those of different structures are separated. The learned distance metric model is then used to compute the similarity measure between image patches in the label fusion. The proposed method has been validated for segmenting hippocampus based on the EADC-ADNI dataset and the experimental results have demonstrated that our method can achieve better segmentation performance than the existing weighted voting label fusion methods with predefined metric models.