Random spatial structure of geometric deformations and Bayesian nonparametrics

Christof Seiler, Xavier Pennec, Susan Holmes

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingAcademicpeer-review


Our work is motivated by the geometric study of lower back pain from patient ct images. In this paper, we take a first step towards that goal by introducing a data-driven way of identifying anatomical regions of interest. We propose a probabilistic model of the geometrical variability and describe individual patients as noisy deformations of a random spatial structure (modeled as regions) from a common template. The random regions are generated using the distance dependent chinese restaurant process. We employ the gibbs sampler to infer regions from a set of noisy deformation fields. Each step of the sampler involves model selection (bayes factor) to decide about fusing regions. In the discussion, we highlight connections between image registration and markov chain monte carlo methods.keywordsmarkov chain monte carlomonte carloimage registrationgibbs samplermarkov chain monte carlo methodthese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Original languageEnglish
Title of host publicationGeometric Science of Information
Number of pages8
Publication statusPublished - 2013
Externally publishedYes

Cite this