Skip to Main Content
Medical image segmentation techniques typically require some form of expert human supervision to provide accurate and consistent identification of anatomic structures of interest. In this paper we briefly explain the traditional wavelet-domain hidden Markov tree (HMT) multi-scale segmentation method and present a multiscale contextual label tree (CLT) method according to the dependency information between image blocks belong to different scales and the algorithm to convert coarse scale into fine-scale. We then illustrate the approach on the segmentation of abdominal organs from MR images and brain structures from CT images. Further study is required to determine whether the proposed algorithm is indeed capable of providing consistently superior segmentation.