Chapter

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2004

Volume 3216 of the series Lecture Notes in Computer Science pp 763-770

Integrated Intensity and Point-Feature Nonrigid Registration

  • Xenophon PapademetrisAffiliated withLancaster UniversityCarnegie Mellon UniversityDepartment of Biomedical Engineering, Yale University New HavenDepartment of Diag. Radiology, Yale University New Haven
  • , Andrea P. JackowskiAffiliated withCarnegie Mellon UniversityChild Study Center, Yale University New Haven
  • , Robert T. SchultzAffiliated withCarnegie Mellon UniversityDepartment of Diag. Radiology, Yale University New HavenChild Study Center, Yale University New Haven
  • , Lawrence H. StaibAffiliated withLancaster UniversityCarnegie Mellon UniversityDepartment of Biomedical Engineering, Yale University New HavenDepartment of Diag. Radiology, Yale University New Haven
  • , James S. DuncanAffiliated withLancaster UniversityCarnegie Mellon UniversityDepartment of Biomedical Engineering, Yale University New HavenDepartment of Diag. Radiology, Yale University New Haven

* Final gross prices may vary according to local VAT.

Get Access

Abstract

In this work, we present a method for the integration of fea- ture and intensity information for non rigid registration. Our method is based on a free-form deformation model, and uses a normalized mu- tual information intensity similarity metric to match intensities and the robust point matching framework to estimate feature (point) correspon- dences. The intensity and feature components of the registration are posed in a single energy functional with associated weights. We com- pare our method to both point-based and intensity-based registrations. In particular, we evaluate registration accuracy as measured by point landmark distances and image intensity similarity on a set of seventeen normal subjects. These results suggest that the integration of intensity and point-based registration is highly effective in yielding more accurate registrations.