¡¡Chinese Journal of Computers   Full Text
  TitleMedical Image Registration Based on Modified Spring-Mass Model
  AuthorsGONG Yong-Yi1),2),3) LUO Xiao-Nan1),2) JIA Wei-Jia4) WONG Hau San4)
  Address1)(Computer Application Institute, Sun Yat-Sen University, Guangzhou 510275)
2)(Key Laboratory of Digital Life of Ministry of Education, Sun Yat-Sen University, Guangzhou 510275)
3)(College of Information Engineering, Zhejiang University of Technology, Hangzhou 310014)
4)(Department of Computation Science, City University of Hong Kong, Kowloon, Hong Kong)
  Year2008
  IssueNo.7(1224¡ª1233)
  Abstract &
  Background 		Abstract Image registration plays an important role in medical image processing. In this paper, a topology preservation image registration algorithm based on modified Spring-Mass model is presented. The image registration problem is transferred to the problem of mechanics balance, by defining the external force as the G¢{teaux derivate on mass displace field of template and source image¡¯ SSD measure. A scheme modifying the mass¡¯s velocity directly is used for the topology preservation of spring-mass model¡¯s evolving. Also, Multi-layers scheme is used for more efficient evolving. A series of experimental results are offered to show the useful of the method.
Keywords image registration; spring-mass model; topology preservation
Background The research group has being conducting numerous experimental researches, such as the numerical simulation research supported by the National Science Fund of Guangdong province (032202), the biological research and computer simulation research based on artificial bone which is supported by the National Science Fund of Guangdong province (04300461) and the Science & Technology Grant (2004B33101005), the medical image segmentation and 3-D re-construction research which is supported by the Group Fund of Guangdong province (20023001).This work is also partly supported by the National Science Fund for Distinguished Young Scholars (grant No.60525213), the Key Laboratory of Multimedia and Intelligent Software (Beijing University of Technology) in Beijing and the Key Project (grant No.60533030) of NSFC, CityU Applied R&D Funding (ARD) (9668009).
Their researches include the image matching of medical images. The authors have found its possibility to keep the topology preservation of deformation field based on Spring-mass model. The authors present a novel scheme just by modifying the velocity of mass to keep the topology preservation of SMM¡¯s evolving. The experiments show their works are useful.