Computational NeuroscienceAnatomical landmarks for registration of experimental image data to volumetric rodent brain atlasing templates
Introduction
A rapid growth of data collection from the rodent brain has taken place over the recent years, strengthened by the extensive use of genetically engineered animals for various types of research. An initiative for providing universal and convenient access to and management of these huge amounts of data has been organized around anatomical reference systems. In 2008, the International Neuroinformatics Coordinating Facility (INCF) launched the Digital Atlasing Program (Bjaalie and Grillner, 2007, Boline et al., 2007), resulting in the development of Waxholm Space (WHS) (Johnson et al., 2010, Hawrylycz, 2009, Bowden et al., 2011) and the associated Digital Atlasing Infrastructure (DAI) (Zaslavsky et al., 2014, Hawrylycz et al., 2011). WHS is a reference space for the rodent brain based on a coordinate system anchored to anatomical landmarks within the brain. Anatomical context is provided by high resolution atlasing templates of multimodal volumetric image material, including MRI, from the brain of the mouse (Johnson et al., 2010) and rat (Papp et al., 2014). The WHS standard serves as a cornerstone of the DAI, a collection of services that provide interoperability between existing and newly arising brain atlases (Allen Brain Atlas, http://www.brain-map.org, Lein et al., 2007; Edinburgh Mouse Atlas Project, http://www.emouseatlas.org/emap, Baldock et al., 2003; Cell Centered Database, http://ccdb.ucsd.edu/index.shtm).
A key step to integrating diverse types of experimental data e.g. gene expression and protein distribution studies, electrophysiology, and functional imaging, is to link the data to a standard anatomical reference, in other words, to register these datasets to a known spatial standard, or atlas. The microscopic resolution Waxholm Space atlas templates provide a basis for spatial registration of diverse types of whole brain data. For MRI datasets, a variety of automated registration methods is available utilizing voxel-based information such as image intensity or contrast (Avants et al., 2011, Denton et al., 1999, Kroon and Slump, 2009, Schnabel et al., 2001; for review see Klein et al., 2009). Such registration methods require images of identical or similar modalities, are sensitive to large-scale distortions, and have limitations for matching anatomical features between brains, in particular between normal and lesioned or diseased brains (Pantazis et al., 2010). Landmark-based registration (Bookstein, 1989, Rohr et al., 2001; for comparison to voxel-based methods, see e.g. Zuk and Atkins, 1996) can be used as an alternative approach, especially when dealing with experimental images of different modalities, with various levels of anatomical detail or resolution with sufficient accuracy for many applications (Gaser et al., 2012, Hsu et al., 2002, Viceic et al., 2009; Weisbecker, 2012; West et al., 2001, Zhang et al., 2006). A landmark is an image feature that serves as a reference: an anatomical characteristic of the brain used for orientation and comparison of data from different experimental animals. When used in registration, landmarks are often reduced to single easily identifiable points to enhance spatial precision. Defining corresponding sets of landmarks in the source and target (reference) datasets determines how the source is mapped to the target. A set of landmarks can be also used for quantifying the accuracy of other registration methods (Castillo et al., 2009, Grachev et al., 1999, Jiang and Johnson, 2011).
The aim of this study was to establish a standardized set of anatomical landmarks in the rodent brain with a primary focus on facilitating registration of experimental datasets to volumetric rodent brain atlasing templates such as the WHS MRI template. As a result, we present 16 landmarks identified with high reliability in MRI of the mouse brain, accompanied by graphical and textual guidelines that allow easy localization of each landmark without particular anatomical experience. We present an example scenario for landmark-based registration to the WHS template based on these landmarks and evaluate the level of precision attained. Further, we successfully identified most landmarks in multiple types of image datasets from the mouse and rat brain representing a range of modalities, image resolutions, and acquisition angles.
Section snippets
Materials and methods
Six adult male C57BL/6J mice (age 24 weeks, weight 30 g, Franz-Penzoldt-Zentrum, Erlangen, Germany) were used in our measurements (Table 1). The animals were handled in accordance with the German Animal Protection Law of 2008 (paragraph 8.1), and all procedures were approved by the Institutional Animal Care and Use Committee of the Regierungspräsidium Mittelfranken.
Results
We defined a set of anatomical landmarks in the mouse brain using an MRI template with three different contrasts from two adult male C57BL/6J mice. We created standardized documentation in both written and graphical form as guidance for locating each landmark in anatomical images of the brain. The documentation, along with landmark positions marked in the mouse WHS template images, is made available to the public via the web-based atlasing tool Scalable Brain Atlas (//sba.incf.org/main/coronal3d.php?template=WHS12%26plugin=Landmarks
Discussion
Integration of various types of data from the brain is the focus of global research programs such as the Human Brain Project, which envisions building a knowledge base of the brain by accumulating and spatially linking imaging and other types of data from the human brain as well as from the brains of experimental animals (HBP Pilot Report, Ailamaki et al., 2012). The framework for linking and managing these large amounts of data is standard anatomical references, or atlases based on
Acknowledgements
We would like to acknowledge the International Neuroinformatics Coordinating Facility (INCF) for financial support of this work (Grant ZZCALINFLO). We thank Johannes Kaesser for technical assistance, animal preparation and scanning. We are grateful to INCF Waxholm Space Task Force members for creative input and suggestions, as well as to all the colleagues from the working group of Andreas Hess (Institute of Experimental and Clinical Pharmacology and Toxicology, FAU Erlangen-Nuremberg) who
References (49)
- et al.
A reproducible evaluation of ANTs similarity metric performance in brain image registration
Neuroimage
(2011) - et al.
A symmetrical Waxholm canonical mouse brain for NeuroMaps
J Neurosci Methods
(2011) - et al.
Neuroanatomical differences between mouse strains as shown by high-resolution 3D MRI
Neuroimage
(2006) - et al.
Deformation-based brain morphometry in rats
Neuroimage
(2012) - et al.
A method for assessing the accuracy of intersubject registration of the human brain using anatomic landmarks
Neuroimage
(1999) - et al.
Microscopic diffusion tensor atlas of the mouse brain
Neuroimage
(2011) - et al.
Waxholm space: an image-based reference for coordinating mouse brain research
Neuroimage
(2010) - et al.
A multidimensional magnetic resonance histology atlas of the Wistar rat brain
Neuroimage
(2012) - et al.
Evaluation of 14 nonlinear deformation algorithms applied to human brain MRI registration
Neuroimage
(2009) - et al.
Variability of bregma in 300 gram Long-Evans and Sprague-Dawley rats
Physiol Behav
(1984)
Comparison of landmark-based and automatic methods for cortical surface registration
Neuroimage
Waxholm Space atlas of the Sprague Dawley rat brain
Neuroimage
Bregma, lambda and the interaural midpoint in stereotaxic surgery with rats of different sex, strain and weight
J Neurosci Methods
Software-guided standardization of manual landmark data in medical images
Z Med Phys
A template for spatial normalisation of MR images of the rat brain
J Neurosci Methods
Variability in the stereotaxic position of cerebral points in the albino rat
Brain Res Bull
Local landmark-based registration for fMRI group studies of nonprimary auditory cortex
Neuroimage
The variability of the interaural line vs the stability of bregma in rat stereotaxic surgery
Physiol Behav
The human brain project: A report to the European commission
EMAP and EMAGE: a framework for understanding spatially organized data
Neuroinformatics
Global neuroinformatics: the international neuroinformatics coordinating facility
J Neurosci
Workshop report: 1st INCF workshop on mouse and rat brain digital atlasing systems
Principal warps – thin-plate splines and the decomposition of deformations
IEEE Trans Pattern Anal Mach Intel
A framework for evaluation of deformable image registration spatial accuracy using large landmark point sets
Phys Med Biol
Cited by (52)
A toolkit for stroke infarct volume estimation in rodents
2024, NeuroImageWhere do you measure the Bregma for rodent stereotaxic surgery?
2023, IBRO Neuroscience ReportsSuper-resolution reconstruction in ultrahigh-field MRI
2023, Biophysical ReportsHomologous organization of cerebellar pathways to sensory, motor, and associative forebrain
2021, Cell ReportsCitation Excerpt :Cell detection parameters were optimized by two users iterating through a set of varying ClearMap detection parameters and selecting those that minimized false positives while labeling only c-Fos positive neurons with high signal-to-noise ratio. To quantify atlas registration accuracy, blinded users labeled readily identifiable points in the PMA (similar to Sergejeva et al., 2015) in four sets of unregistered, affine-only registered, and fully registered volumes (Figure 1H; Figure S2I). This allowed for quantification of landmark distances (Sergejeva et al., 2015) between the PMA and brains at different stages of registration.
An analytical workflow for seed-based correlation and independent component analysis in interventional resting-state fMRI studies
2021, Neuroscience ResearchCitation Excerpt :Then, the motion parameters and the noise components of all filtered datasets from MELODIC were automatically classified and regressed by FSL/FIX at a threshold of 20 (Salimi-Khorshidi et al., 2014, Griffanti et al., 2014). The de-noised fMRI images for each session were then co-registered to their respective T2-weighted images using six-parameter rigid body registration using FLIRT (FSL Linear Image Registration Tool; Jenkinson and Smith, 2001, Jenkinson et al., 2002) and normalised to a Sprague Dawley brain atlas (Papp et al., 2014, Kjonigsen et al., 2015, Sergejeva et al., 2015) with nine degrees of freedom registration. The atlas was first down-sampled by a factor of eight to better match the voxel size of the 4D functional data.
- 1
These authors contributed equally.