A non-rigid registration method for mouse whole body skeleton registration Di Xiao* a , David Zahra b , Pierrick Bourgeat a , Paula Berghofer b , Oscar Acosta Tamayo a , Catriona Wimberley b , Marie Claude Gregoire b , Olivier Salvado a a The Australian e-Health Research Center, ICT, CSIRO, Australia b Radiopharmaceutical Research Institute, ANSTO, Australia ABSTRACT Micro-CT/PET imaging scanner provides a powerful tool to study tumor in small rodents in response to therapy. Accurate image registration is a necessary step to quantify the characteristics of images acquired in longitudinal studies. Small animal registration is challenging because of the very deformable body of the animal often resulting in different postures despite physical restraints. In this paper, we propose a non-rigid registration approach for the automatic registration of mouse whole body skeletons, which is based on our improved 3D shape context non-rigid registration method. The whole body skeleton registration approach has been tested on 21 pairs of mouse CT images with variations of individuals and time-instances. The experimental results demonstrated the stability and accuracy of the proposed method for automatic mouse whole body skeleton registration. Keywords: 3D non-rigid registration, mouse whole body skeleton registration, 3D shape context model 1. INTRODUCTION Small animal imaging is increasingly used as a pre-clinical tool to identify new imaging agent, or assess effectiveness of therapy of diseases through their MRI, micro-CT and Position Emission Tomography (PET) images in vivo. This involves scanning a cohort of small rodents (typically mice and rats), and computing population statistics of specific organs or measuring and quantifying temporal changes in a region of interest (ROI). One challenging step to study a large numbers of individual animals is performing spatial normalization before any subsequent processing. Common tasks include the tracking of tumor size variations and shape in a longitudinal experiment of rats or mice CT/PET images acquired over time or mapping specific organs from an atlas to any new scanned images. For small animals, because of the articulated joints and their anatomical structures, it is challenging to position the animals in a same position with a same posture for each scan. The use of physical support for the animals reduces large posture differences but a significant deformation of the animal body still exists from one scan to another and between different animals. The only easily identifiable and robust anatomical features present in CT images are the skeletons, lungs and skin. Recently, a few methods have been proposed to address the automatic registration of small animal whole body skeletons. Baiker et al. proposed an automatic articulated registration method for mouse skeleton registration by identifying joints and individual bones and traversing a hierarchical mouse skeleton tree predefined and registering each part by iterative closest point (ICP) from a coarse to fine method [1]. Li et al. [2] proposed using robust point-based registration [3] and softassign algorithm [4] for mouse whole body skeleton registration in their two-step registration process and tested it on a mouse’s serial CT images. Hesheng et al. [5] proposed a deformable image registration method, consisting of a global affine transformation and a local B-splines deformation, for mouse whole body skeleton registration. The deformation model was incorporated into robust point matching (TPS-RPM) [3] method for estimating point correspondences and surface deformation. The first method is a combination of piecewise rigid registration methods for automatically labeling of whole body skeletons. The later two methods aim at a whole body registration by non-rigid registration method without skeleton structure labeling. *Di.Xiao@csiro.au; phone 61 2 97179893; http://www.aehrc.com/biomedical_imaging/ Medical Imaging 2010: Biomedical Applications in Molecular, Structural, and Functional Imaging, edited by Robert C. Molthen, John B. Weaver, Proc. of SPIE Vol. 7626, 76261N · © 2010 SPIE CCC code: 1605-7422/10/$18 · doi: 10.1117/12.844067 Proc. of SPIE Vol. 7626 76261N-1 Downloaded From: http://proceedings.spiedigitallibrary.org/ on 12/02/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx