Multi-modal image set registration and atlas formation

Abstract

In this paper, we present a Bayesian framework for both generating inter-subject large deformation transformations between two multi-modal image sets of the brain and for forming multi-class brain atlases. In this framework, the estimated transformations are generated using maximal information about the underlying neuroanatomy present in each of the different modalities. This modality independent registration framework is achieved by jointly estimating the posterior probabilities associated with the multi-modal image sets and the high-dimensional registration transformations mapping these posteriors. To maximally use the information present in all the modalities for registration, Kullback-Leibler divergence between the estimated posteriors is minimized. Registration results for image sets composed of multi-modal MR images of healthy adult human brains are presented. Atlas formation results are presented for a population of five infant human brains.

Real-Time 2D/3D Deformable Registration Using Metric Learning

“…REALMS computes a respiratory Fréchet mean image J from the RCCT dataset via an LDDMM (Large Deformation Diffeomorphic Metric Mapping) framework described in Lorenzen et al [11]. The Fréchet mean J, as well as the diffeomorphic deformations φ from the mean J to each image J τ , are computed using a fluidflow distance metric:…”

Image registration and atlas-based segmentation of cardiac outflow velocity profiles

“…The arbitrary deformed real image can be used for the registration evaluation if the manually segmented original (non-deformed) image is used as a ground-truth. Finally last criteria in the registration evaluation process was the inverse consistency as proposed by Christensen et al [38] and Lorenzen et al [32].…”

“…Although inverse consistency does not guarantee the accuracy of the registration, it is often preferable or even used as measure of quality of the registration [38,32]. This, along with the desire to quantify the bi-directional transformation error, are the main reasons for the additional validation using the inverse consistency test [43,44].…”

“…Atlases have broad application in medical image segmentation and registration and are often used in computer aided diagnosis to measure the shape of an object or detect the morphological differences between patient groups. Various techniques for atlas construction are developed for different human organs, like the heart [25,26,27] and especially the brain [28,29,30,31,32,33,34,35,36,37]. In this paper we use a statistical model as an atlas and an in-silico phantom model for evaluation.…”

What is where and why it is important

“…These atlases may even have a time-varying component (Janke et al, 2001), allowing subjects of different ages to be brought into the atlas using an age-appropriate transformation. Rather than average images together voxel-by-voxel to produce a blurred template, as was done in many 'first generation' statistical atlases, many groups are developing practical methods to create well-resolved canonical atlas images that represent the statistical mean anatomy for a population, using deformation averaging (Collins et al, 1995); (Thompson et al, 2000); (Kochunov P, 2002); (Twining et al, 2005)), Lie group methods on deformation tensors (Woods, 2003); (Lepore et al, 2006)), or geodesics on groups of diffeomorphic flows (Joshi et al, 2004); (Miller et al, 2005); (Lorenzen et al, 2006)). These approaches are complex, but are advantageous as they are close (in a strictly defined mathematical sense) to the brains being normalized to them and are likely to improve spatial accuracy and reduce sources of bias when comparing datasets in a canonical coordinate system.…”