Abstract
Publicly available scientific resources help establish evaluation standards, provide a platform for teaching and improve reproducibility. Version 4 of the Insight ToolKit (ITK4) seeks to establish new standards in publicly available image registration methodology. ITK4 makes several advances in comparison to previous versions of ITK. ITK4 supports both multivariate images and objective functions; it also unifies high-dimensional (deformation field) and low-dimensional (affine) transformations with metrics that are reusable across transform types and with composite transforms that allow arbitrary series of geometric mappings to be chained together seamlessly. Metrics and optimizers take advantage of multi-core resources, when available. Furthermore, ITK4 reduces the parameter optimization burden via principled heuristics that automatically set scaling across disparate parameter types (rotations vs. translations). A related approach also constrains steps sizes for gradient-based optimizers. The result is that tuning for different metrics and/or image pairs is rarely necessary allowing the researcher to more easily focus on design/comparison of registration strategies. In total, the ITK4 contribution is intended as a structure to support reproducible research practices, will provide a more extensive foundation against which to evaluate new work in image registration and also enable application level programmers a broad suite of tools on which to build. Finally, we contextualize this work with a reference registration evaluation study with application to pediatric brain labeling.1
Highlights
While Advanced Normalization Tools (ANTs) still serves as intermediate accesspoint to these tools, it provides a high-degree of customization possibilities through a command line interface and scripting
ANTs was instrumental to a first-place finish in SATA 2013 in two of three categories where the ANTs approach was considerably simpler than that employed by close finishers
The ANTs solution was the only one that was fully automated resulting in a ≈15% performance loss which can be overcome by a modicum of user intervention
Summary
As image registration methods mature—and their capabilities become more widely recognized—the number of applications increase (Rueckert et al, 1999; van Dalen et al, 2004; Miller et al, 2005; Shelton et al, 2005; Chen et al, 2008; Baloch and Davatzikos, 2009; Cheung and Krishnan, 2009; Peyrat et al, 2010; Fedorov et al, 2011; Kikinis and Pieper, 2011; Metz et al, 2011; Murphy et al, 2011). Despite the increasing relevance of image registration across application domains, there are relatively few reference algorithm implementations available to the community. The BRAINSFit algorithm is integrated into Slicer for user-guided registration (Kikinis and Pieper, 2011) Each of these toolkits has both strengths and weaknesses (Klein et al, 2010; Murphy et al, 2011) and was enabled by an ITK core
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.