Surgical Planning Laboratory - Brigham & Women's Hospital - Boston, Massachusetts USA - a teaching affiliate of Harvard Medical School

Surgical Planning Laboratory

The Publication Database hosted by SPL

All Publications | Upload | Advanced Search | Gallery View | Download Statistics | Help | Import | Log in

Gaussian Process Interpolation for Uncertainty Estimation in Image Registration

Institution:
1Computer Science and Artificial Intelligence Lab, Massachusetts Institute of Technology, Cambridge, MA, USA.
2Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA.
3Surgical Planning Laboratory, Brigham and Women's Hospital and Harvard Medical School, Boston, MA, USA.
Publisher:
Int Conf Med Image Comput Comput Assist Interv. MICCAI 2014
Publication Date:
Sep-2014
Journal:
Med Image Comput Comput Assist Interv
Volume Number:
17
Issue Number:
Pt 1
Pages:
267-74
Citation:
Int Conf Med Image Comput Comput Assist Interv. 2014 Sep;17(Pt 1):267-74.
PubMed ID:
25333127
PMCID:
PMC4219919
Appears in Collections:
NA-MIC, NAC, NCIGT, SPL
Sponsors:
P41 EB015898/EB/NIBIB NIH HHS/United States
P41 EB015902/EB/NIBIB NIH HHS/United States
U54 EB005149/EB/NIBIB NIH HHS/United States
Generated Citation:
Wachinger C., Golland P., Reuter M., Wells III W.M. Gaussian Process Interpolation for Uncertainty Estimation in Image Registration. Int Conf Med Image Comput Comput Assist Interv. 2014 Sep;17(Pt 1):267-74. PMID: 25333127. PMCID: PMC4219919.
Downloaded: 1177 times. [view map]
Paper: Download, View online
Export citation:
Google Scholar: link

Intensity-based image registration requires resampling images on a common grid to evaluate the similarity function. The uncertainty of interpolation varies across the image, depending on the location of resampled points relative to the base grid. We propose to perform Bayesian inference with Gaussian processes, where the covariance matrix of the Gaussian process posterior distribution estimates the uncertainty in interpolation. The Gaussian process replaces a single image with a distribution over images that we integrate into a generative model for registration. Marginalization over resampled images leads to a new similarity measure that includes the uncertainty of the interpolation. We demonstrate that our approach increases the registration accuracy and propose an efficient approximation scheme that enables seamless integration with existing registration methods.

Additional Material
1 File (91.63kB)
Wachinger-MICCAI2014-fig1.jpg (91.63kB)