Segmentation of the C57BL/6J mouse cerebellum in magnetic resonance images

• DOI: 10.1016/j.neuroimage.2012.05.061
• Publication Date: 2012-05-30T22:16:23Z
• Authors: Jeremy F.P. Ullmann; Marianne D. Keller; Charles Watson; Andrew L. Janke; Nyoman D. Kurniawan; Zhengyi Yang; Kay Richards; George Paxinos; Gary F. Egan; Steven Petrou; Perry Bartlett; Graham J. Galloway; David C. Reutens
• Keywords: 2805 Cognitive Neuroscience; 610; magnetic resonance; Mice; 03 medical and health sciences; Segmentation; 0302 clinical medicine; probabilistic map; Cerebellum; 616; Image Processing, Computer-Assisted; Animals; atlas; Vermis; Brain Mapping; Mouse brain; segmentation; vermis; Magnetic Resonance Imaging; Mice, Inbred C57BL; Probabilistic map; Magnetic resonance; mouse brain; 2808 Neurology; Atlas

The C57BL mouse is the centerpiece of efforts to use gene-targeting technology to understand cerebellar pathology, thus creating a need for a detailed magnetic resonance imaging (MRI) atlas of the cerebellum of this strain. In this study we present a methodology for systematic delineation of the vermal and hemispheric lobules of the C57BL/6J mouse cerebellum in magnetic resonance images. We have successfully delineated 38 cerebellar and cerebellar-related structures. The higher signal-to-noise ratio achieved by group averaging facilitated the identification of anatomical structures. In addition, we have calculated average region volumes and created probabilistic maps for each structure. The segmentation method and the probabilistic maps we have created will provide a foundation for future studies of cerebellar disorders using transgenic mouse models.