RABBIT: Rapid Alignment of Brains by Building Intermediate Templates

By Songyuan Tang, Yong Fan, Guorong Wu, Minjeong Kim, Dinggang Shen. Medical Imaging, 2009

416 citations recorded by the source corpus.

This paper proposes a brain image registration algorithm, called RABBIT, which achieves fast and accurate image registration by using an intermediate template generated by a statistical shape deformation model during the image registration procedure. The statistical brain shape deformation information is learned by means of principal component analysis (PCA) from a set of training brain deformations, each of them linking a selected template to an individual brain sample. Using the statistical deformation information, the template image can be registered to a new individual image by optimizing a statistical deformation model with a small number of parameters, thus generating an intermediate template very close to the individual brain image. The remaining shape difference between the intermediate template and the individual brain is then minimized by a general registration algorithm, such as HAMMER. With the help of the intermediate template, the registration between the template and individual brain images can be achieved fast and with similar registration accuracy as HAMMER. The effectiveness of the RABBIT has been evaluated by using both simulated atrophy data and real brain images. The experimental results show that RABBIT can achieve over five times speedup, compared to HAMMER, without losing any registration accuracy or statistical power in detecting brain atrophy.

Subfields: Computer Vision and Pattern Recognition, Neurology, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/455eea02c09685dc47855257771d8213680e8a61

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Neurology

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Effective connectivity: Influence, causality and biophysical modeling (2011, methodology): “…for modeling and ﬁ tting systems de ﬁ ned by differential equations in bioinformatics is “ Potters Wheel ” (Maiwald and Timmer, 2008), which uses a pro ﬁ le likelihood approach (Raue et al., 2009) to explore “ practicalIdenti ﬁ ability ” inadditiontostructural(theoretical)Identi ﬁ ability.”
Hierarchical feature representation and multimodal fusion with deep learning for AD/MCI diagnosis (2014, neutral): “Accepted 28 June 2014 Available online 18 July 2014”
Advancing functional connectivity research from association to causation (2019, background): “…connectivity (although attempts have been made to address this 1,2 ), with emphasis sometimes placed on the target theoretical property of whether a connection is direct vs. indirect (e.g., via a third brain region) 7,26,27 or whether a connection is directed vs. undirected/bidirectional 2,7,28 .”
Handbook of Big Data (2016, background): “The framework, discussed in this chapter, provides a pathway from our scientiﬁc aim to estimation of a statistical parameter that best approximates our causal eﬀect, while keeping any assumptions transparent.”
Handbook of Functional MRI Data Analysis: Index (2011, methodology): “To address this issue, Ramsey et al. (2010) developed a method called IMaGES (independent multiply assessed greedy equivalence search), which performs a search across multiple subjects, even if those subjects do not share all of the same variables (e.g., if one subject has no significant signal…”
Causal inference in partially linear structural equation models (2016, background): “Examples include the identification of causal molecular mechanisms in genomics [24, 25], the investigation of causal relations among activity in brain regions from fMRI data [19] or the search for causal associations in public health [7].”
A million variables and more: the Fast Greedy Equivalence Search algorithm for learning high-dimensional graphical causal models, with an application to functional magnetic resonance images (2016, methodology): “For that reason, the IMaGES algorithm [14] runs each step of GES separately on each data set and updates the inferred MEC by adding (or in the backwards phase, deleting) the edge whose total score summed over the scores from the several data sets is best.”
Functional and Effective Connectivity: A Review (2011, neutral): “This degeneracy has been highlighted by Ramsey et al. (2010) in the setting of effective connectivity analysis.”
The Domain of Organizational Cognitive Neuroscience (2012, background): “However, as has been shown previously, this is far from guaranteed and may be impossible to prove (e.g., Price & Friston, 1997; Ramsey et al., 2010).”
Deformable Medical Image Registration: A Survey (2013, neutral): “[172] also used PCA to learn an SDM to accelerate image alignment.”
Learning equivalence classes of acyclic models with latent and selection variables from multiple datasets with overlapping variables (2011, result): “Some edges observed are also consistent with the results in Ramsey et al. [2010], which analyzed this data using a different method.”
Score-based causal learning in additive noise models (2013, background): “Some real-life examples include identifying gene expression networks [4,5] and analysing fMRI data from the human brain.[6] When modelling causal networks between some given variables, structural equation models (SEMs) are used frequently, where each variable is expressed as a function of some other”
Group search algorithm recovers effective connectivity maps for individuals in homogeneous and heterogeneous samples (2012, methodology): “On the one hand, true heterogeneity in individual processes may make group-level findings erroneous because the resulting map may not apply to any one individual (Miller and van Horn, 2007; Molenaar, 2004; Ramsey et al., 2010).”
The future of FMRI connectivity (2012, methodology): “Such methods may look at the probability of pairs of variables conditional on sets of other variables; for example, Bayes net methods (Ramsey et al., 2010) in general estimate directionality by first orienting “unshielded colliders” (paths of the form A→B←C, where a node is fed into by the others)…”
Nonlinear directed acyclic structure learning with weakly additive noise models (2009, methodology): “We also ran kPC on data from an fMRI experiment that is analyzed in [21] where nonlinear dependencies can be observed.”
Latent feature representation with stacked auto-encoder for AD/MCI diagnosis (2013, neutral): “…other advanced registration methods can also be applied for this process (Friston et al. 1995; Evans and Collins 1997; Rueckert et al. 1999; Shen et al. 1999; Wu et al. 2006; Xue et al. 2006a, b; Avants et al. 2008; Yang et al. 2008; Tang et al. 2009; Vercauteren et al. 2009; Jia et al. 2010).”
Network modelling methods for FMRI (2011, neutral): “Such methods may look at the probability of pairs of variables conditional on sets of other variables; for example, Bayes net methods (Ramsey et al., 2010) in general estimate directionality by”