A 3D HIGH RESOLUTION MRI ATLAS OF THE SHEEP BRAIN
Arsene Longin Ella1, José Delgadillo1, Philippe Chemineau1, and Matthieu Keller1

1Laboratory of Reproductive Physiology and Behavior, INRA - Centre Val de Loire, Nouzilly, France

Synopsis

The sheep model was first used in the field of reproductive physiology in agronomy to improve milk and meat production, and then was brought into fundamental and preclinical neurosciences. Since a decade, MR studies performed on this model are increasingly reported. To play an important role in MR translational neuroscientific research, a brain template and an atlas are therefore necessary. Simultaneously, two MR templates were proposed in 2015. To complete the set of MR tools, we computed a high resolution 3D in-vivo sheep brain atlas including: i) gyri and sulci ii) inner structures iii) main external structures.

INTRODUCTION

The sheep model was first used in the field of reproductive physiology in agronomy to improve the production of milk and meat1, and then was brought into a broad range of fundamental and preclinical neuroscience studies.2 Since a decade, MR studies performed on this model are increasingly reported.3,4 To play an important role in MR translational neuroscientific research, a template image and an atlas of the sheep brain are therefore necessary. Simultaneously, two MR templates were proposed in 20155,6 and to complete the set of MR tools, in this study, we computed a high resolution 3D in-vivo sheep brain atlas including: i) gyri and sulci, ii) inner structures, iii) main external structures (cerebrum, brain stem and cerebellum).

MATERIALS AND METHODS

MRI scans were acquired on anesthetized 18 female sheep of 4 years old (with 20 mg/kg of ketamine, intubated and maintained on 3% isoflurane vaporized in oxygen) with a 3T VERIO Siemens device at the CIRE platform of INRA Nouzilly France. Parameters were for the Magnetization Prepared Rapid Gradient Echo (MPRAGE) T1W: TR 2500 ms; TE 3.6 ms; TI 900; FA 12°; NEX 4; matrix 384×384; FOV192 mm; 288 sagittal slices with a thickness of 0.5 mm; and for the SPACE T2W: TR 4000 ms; TE 43 ms; FA 120°; NEX 5; matrix 512×512; FOV180 mm; 208 sagittal slices with a thickness of 0.35 mm, resampled to a voxel size of 0.5x0.5x0.5 mm3. T1W and T2W templates images and non-linear tissue probability maps (TMP) where computed using an iterative optimized non-linear registration in FSL (FMRIB Analysis Group, Oxford, UK) as described in 6,7. Finally, atlas structures where extracted either by segmenting templates using FSL-FAST or by an intensity threshold of TPM and identified.

RESULTS

Results showed a high spatial resolution of templates and tissue probability maps (TPM) of the gray matter (GM), the white matter (WM) and the cerebrospinal fluid or CSF (fig.1) which led to the segmentation of 20 gyri (fig.2) and 21 deep internal structures (fig.3). We were also able to identify the main external structures (fig.4): cerebrum, cerebellum, pons, pituitary gland, medulla oblongata, and midbrain.

DISCUSSION AND CONCLUSION

If a single intensity threshold of TPM of the CSF, WM, and GM allowed to retrieve structures like ventricles (CSF), corpus callosum (WM) and striatal complex (GM), a mix of both FSL-FAST segmentation of templates and TPM intensity threshold was necessary to overcome the delineation of smaller structures with a minimum of manual interaction. This was helped by the high spatial resolution presented by templates and TPM where those small structures were already visible. The atlas presented in this study, associated to early developed templates provide a full set of MRI tools necessary to perform group studies for different types of MR imaging techniques and may be a key step for the use of the sheep model in the translational brain research.

Acknowledgements

No acknowledgement found.

References

1. Delgadillo JA, Gelez H, Ungerfeld R, et al. The 'male effect' in sheep and goats-Revisiting the dogmas. Behav Brain Res. 2009; 200(2):304–14.

2. Brus M, Meurisse M, Gheusi G, et al. Dynamics of olfactory and hippocampal neurogenesis in adult sheep. J Comp Neurol. 2013; 521(1):169–88.

3. Opdam HI, Federico P, Jackson GD, et al. A sheep model for the study of focal epilepsy with concurrent intracranial EEG and functional MRI. Epilepsia. 2002; 43(8):77–87.

4. Gierthmuehlen M, Wang X, Gkogkidis A, et al. Mapping of sheep sensory cortex with a novel microelectrocorticography grid. J Comp Neurol. 2014; 522(16):3590–608.

5. Nitzsche B, Frey S, Collins LD, et al. A stereotaxic, population-averaged T1w ovine brain atlas including cerebral morphology and tissue volumes. Front Neuroanat. 2015; 9(69):1–14.

6. Ella A, Keller M. Construction of an MRI 3D high resolution sheep brain template. Magn Reson Imaging. 2015; http://dx.doi.org/10.1016/j.mri.2015.09.001

7. Ella A & Keller M. Proceedings of the ISMRM 2015.

Figures

Figure 1: Axial (left), coronal (middle) and sagittal (right) slices of GM, WM and CSF prior probability maps computed during non-linear registrations.

Figure 2: Sagittal (right), coronal (middle) and axial (left) slices of (A) the T1W sheep brain template, (B) brain gyri and (C) 3D surface renderings of brain gyri.

Figure 3: Sagittal (right), coronal (middle) and axial (left) slices of (A) T1W and (B) T2W sheep brain templates, and (C) segmented inner brain structures overlapped the T2W template

Figure 4: (A) sheep brain sagittal slice with main brain structures and (B, C, D) different views of their 3D surface renderings.



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
4439