Introduction

Recently, neuroscience research has been extended from non-human primate and rodent models to the domestic dog, who exhibits similar emotion and social processing as human¹. Furthermore, the domestic dog is potentially a promising animal model by genetic manipulations for understanding neural mechanism of psychiatry diseases such as autism. MRI/fMRI as one of the powerful tools can non-invasively explore anatomic variation and functional activation. Luckily, several groups published population-averaged canine brain templates^{2, 3, 4, 5}. However, the dogs they used were greatly different in age, skull shape, body size, and brain anatomy, besides there is yet no high-resolution brain template for purebred dogs, which is essential for understanding brain development and function. In this study, high-resolution T1w and T2w brain templates for male adult purebred beagle would be created.

Methods

Ten purebred adult beagles (all males, weight 10.54 ± 0.9 kg, age 2.38 ± 0.85 years) were scanned at Siemens prisma^fit 3.0T MR scanner (Siemens healthnieer, Erlangen, Germany) using home-made 4 channel Tx/Rx coil. Each dog was pre-anesthetized with intramuscular atropine (0.05 mg/kg), followed by Zoletil 50 (10 mg/kg) and Sumianxin (1.6 mg/kg). Complementary anesthesia was performed every one hour with Zoletil 50 (5 mg/kg). T1 weighted, T2 weighted and proton density weighted images in a 3D volume with a field of view of 128 x 128 x 128mm³ and 0.5mm isotropic voxel size were acquired with MPRAGE (magnetization prepared rapid acquisition gradient echoes), T2 space and GRE, respectively. All the experimental protocols were approved by the local institute Animal Care and Committee. Preprocessing was performed using SPM12 (https://www.fil.ion.ucl.ac.uk/spm/software/spm12/), implemented in Matlab (The Mathworks Inc., USA) and FSL (Functional MRI of the brain Software Library). Initially, subjects were re-orientated to AC-PC space and resampled from 0.5 to 0.25 mm isotropic. Then using FSL fast tools to segment T1w image to initial gray matter (GM), white matter (WM), and cerebral spinal fluid (CSF) tissue images, and using SPM to create tissue probability maps (TPM) template. Fig. 1 exhibits the detailed preprocessing steps. In step 4, individual morphological parameters, including GM, WM, and CSF volumes, were extracted based on CAS-Beagle brain template.

Results

Fig. 2 demoes a typical beagle dog high-resolution skull-stripped T1w brain image and segmented GM, WM and CSF tissue images in each row. T1w image shows high contrast between GM and WM, clear boundary among GM, WM and CSF. High-resolution purebred beagle T1w and T2w brain templates named as CAS-Beagle Template were created. Fig. 3 shows the sagittal, coronal, and transversal CAS-Beagle TPM overlaid on T1w template, which was generated from 10 male adult beagles, whose brain size are 55.7 ± 2.4mm in the right-left dimension (RL), 81.0 ± 2.6mm in the anterior-posterior dimension (AP) and 45.1 ± 1.8mm in the dorsal-ventral dimension (DV). Multiple-transversal slices of CAS-Beagle TPM overlaid on T1w template are presented in Fig. 4. A 21 parcel atlas⁵ were normalized to CAS-Beagle TPM and overlaid on template. The descriptive statistics of beagle brain morphology is listed in Fig. 5. GM volume shows significant correlation only with brain size of DV (r² = 0.7652, p < 0.05, Fig. 5A), the correlation coefficients among other brain morphological parameters are not significant. Only GM volume shows close correlation with body weight (r² = 0.3067, p < 0.05, Fig. 5B), the correlation coefficients between other brain morphological parameters and body weight are not significant. The absolute volumes of GM, WM, CSF, and WB (whole brain) are 48.10 ± 6.50 mL, 23.47 ± 3.00 mL, 2.48 ± 0.72 mL, and 74.05 ± 8.84 mL, respectively. The mean volumes are very close to that of a previous report, but the standard deviations are smaller than that in the report⁵. The ratio of GM to WM volume is 2.06 ± 0.28 and GM to WB volume is 0.65 ± 0.002. Tissue volumes (per kg body weight) are 4.60 ± 0.53 mL for GM, 2.25 ± 0.29 mL for WM, 0.24 ± 0.07 mL for CSF, and 7.09 ± 0.76 mL for WB.

Discussion & Conclusion

High-resolution purebred beagle MRI T1w and T2w templates were successfully created from 10 adult beagles. This template could potentially provide more insight into the promotion of beagle brain template used in emotional and social processing research on human beings and canines. The CAS-Beagle Template would facilitate future studies to answer important questions, such as the variance of cortical topographies and the longitudinal trajectories of brain development along different ages of different breeds of canines.

Acknowledgements

This work was supported in part by the Ministry of Science and Technology of China (2015CB351701), the National Natural Science Foundation of China (31730039, 81871350), National Major Scientific Instruments and Equipment Development Project (ZDYZ2015-2) and Chinese Academy of Sciences Strategic Priority Research Program B grants (XDBS01000000).