Abstract
CT image registration technique is important in the medical image processing which also can be seen as signal processing. But the process of biological mechanics model of energy function is a tedious calculation and longer time-consuming, which cannot meet the requirements of clinical application. In order to solve this problem, through the level set method, this paper solves biological mechanics model of numerical solution of energy function, and the process of CUDA acceleration, so as to improve execution efficiency of medical image registration algorithm based on biological mechanics model and make it meet the clinical application of the real-time requirement. Experiments show that the method can be accurate and fast to abdominal CT image registration, and it also can satisfy the real-time requirement of the image registration clinically.
Similar content being viewed by others
References
Blume M, Navab N, Rafecas M (2012) Joint image and motion reconstruction for PET using a B-spline motion model. Phys Med Biol 57(24):8249–8270
Cao GG, Rong CC, Luo LM (2008) Medical image registration based on differential evolution. J Appl Sci 26(3):274–280
Michalek J, Capek M, Kubinova L (2011) Nonrigid registration of CLSM images of physical sections with discontinuous deformations. Microsc Microanal 17(6):923–936
Persson PO, Strang G (2004) A simple mesh generator in MATLAB. SIAM Rev 46(2):329–345
Sean G, Purang A, Gabor F, Jonathan B, David P, Dan B, Parvin M (2012) Biomechanically constrained groupwise ultrasound to CT registration of the lumbar spine. Med Image Anal 16(3):662–674
Sher SO, Sethian J (1988) Fronts propagating with curvature dependent speed: algorithms based on a Hamilton-Jacobi formulation. Commun Phys 79:12–49
Wang YH, Wang MC, Zhan XD (2003) Reconstruction of liver surface from CT-image. Chin J Med Imaging Technol 19(6):675–676
Wang Z-y, Jian-wei Z, Zhang-can H (2007) Image registration model based on evolutionary modeling. Comput Eng Appl 43(35):71–73
Acknowledgments
Research and application of pancreatic cancer early diagnosis algorithm based on dynamic biomechanical models, School Basic Scientific Research Business Expenses for Northeastern University (N130317004). Research on big data query technology in cloud environment and its application in medical big data, Doctoral Research Fund for Liaoning province (20141011).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Wang, B., Ying, C. Liver medical image registration based on biomechanical model. Multimed Tools Appl 76, 19927–19944 (2017). https://doi.org/10.1007/s11042-016-3828-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-016-3828-x