Robust methods to create ex vivo minimum deformation atlases for brain mapping

• Published in: Methods (San Diego, Calif.) Vol. 73; pp. 18 - 26
• Main Authors: Janke, Andrew L., Ullmann, Jeremy F.P.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.02.2015
• Subjects: Animals; Brain - anatomy & histology; Brain - physiology; Brain atlas; Brain Mapping - methods; Brain Mapping - trends; Ex vivo imaging; Humans; Image Processing, Computer-Assisted - methods; Image Processing, Computer-Assisted - trends; Imaging, Three-Dimensional - methods; Imaging, Three-Dimensional - trends; Magnetic Resonance Imaging - methods; Magnetic Resonance Imaging - trends; Minimum deformation atlas; Registration; Brain atlas; Ex vivo imaging; Registration; Minimum deformation atlas
• ISSN: 1046-2023; 1095-9130
• DOI: 10.1016/j.ymeth.2015.01.005

Highly detailed ex vivo 3D atlases of average structure are of critical importance to neuroscience and its current push to understanding the global microstructure of the brain. Multiple single slice histology sections can no longer provide sufficient detail of inter-slice microstructure and lack out of plane resolution. Two ex vivo methods have emerged that can create such detailed models. High-field micro MRI with the addition of contrast media has allowed intact whole brain microstructure imaging with an isotropic resolution of 15μm in mouse. Blockface imaging has similarly evolved to a point where it is now possible to image an entire brain in a rigorous fashion with an out of plane resolution of 10μm. Despite the destruction of the tissue as part of this process it allows a reconstructed model that is free from cutting artifacts. Both of these methods have been utilised to create minimum deformation atlases that are representative of the respective populations. The MDA atlases allow us unprecedented insight into the commonality and differences in microstructure in cortical structures in specific taxa. In this paper we provide an overview of how to create such MDA models from ex vivo data.