Study of three-dimensional PET and MR image registration based on higher-order mutual information

Study of three-dimensional PET and MR image registration based on higher-order mutual information 1 first-author 2 1 1 corresp Corresponding author E-Mail: wkwu@public.bta.net.cn Corresponding author E-Mail: wkwu@public.bta.net.cn wkwu@public.bta.net.cn Department of Nuclear Medicine, Cancer Hospital, Chinese Academy of Medical Sciences. Peking Union Medical College, Beijing 100021 Department of Nuclear Medicine, Cancer Hospital, Chinese Academy of Medical Sciences. Peking Union Medical College, Beijing 100021 Department of Biomedical Engineering, Capital University of Medical Sciences, Beijing 100054 Department of Biomedical Engineering, Capital University of Medical Sciences, Beijing 100054 Mutual information has currently been one of the most intensively researched measures. It has been proven to be accurate and effective registration measure. Despite the general promising results, mutual information sometimes might lead to misregistration because of neglecting spatial information and treating intensity variations with undue sensitivity. In this paper, an extension of mutual information framework was proposed in which higher-order spatial information regarding image structures was incorporated into the registration processing of PET and MR. The second-order estimate of mutual information algorithm was applied to the registration of seven patients. Evaluation from Vanderbilt University and our visual inspection showed that sub-voxel accuracy and robust results were achieved in all cases with second-order mutual information as the similarity measure and with Powell's multi-dimensional direction set method as optimization strategy. Higher-order mutual information Multi-modality medical image Registration References Thurfjell L , Lau Y H , Andersson J L et al . Eur J Nucl Med , 2000 , 27 : 847 - 856 Eur J Nucl Med Viola P , Wells W M III . Proceedings of 5th International Conference on Computer Vision , Cambridge, MA , 1995 , 16 - 23 Wells WMIII , Viola P , Atsumi H et al . Med Image Anal , 1996 , 1 : 35 - 51 Med Image Anal Maes F , Collignon A , Vandermeulen D et al . IEEE Trans Med Imag , 1997 , 16 : 187 - 198 IEEE Trans Med Imag Yang H , Ma B R , Ren H P et al . Chin J Med Phys (in Chinese) , 2001 , 18 : 69 - 73 Chin J Med Phys (in Chinese) Pluim J P W , Maintz J B A , Viergever M A. IEEE Trans Med Imag , 2000 , 19 : 809 - 814 IEEE Trans Med Imag Rueckert D , Clarkson M J , Hill DLG et al . Proceedings of SPIE Medical Imaging 2000, Image Processing , San Diego, CA , 2000 , 438 - 447 West J , Fitzpatrick J M , Wang M Y et al . J Comput Assist Tomogr , 1997 , 21 : 554 - 566 J Comput Assist Tomogr R318.04 The images and the standard transformation were provided as part of the project, "Retrospective Image Registration Evaluation" (National Institutes of Health, 1 R01 CA89323), the principal investigator. J. Michael Fitzpatrick, Vanderbilt University, Nashville, TN. 2001-12-17 2002-05-01 2021-02-05 2002 Nuclear Science and Techniques 2 13