A High-Resolution 3D In Vivo Rabbit Brain Template from 5T MRI
Poster No:
1495
Submission Type:
Abstract Submission
Authors:
Zhanbo Zhang1,2, Tianhao Zhou1,2, Yong Yang1,2, Chenfei Cao1,2, Xiaodong Zhang1,2, Pengcheng Li1,2, Feng Yue1,2, Junjie Zhuo1,2
Institutions:
1School of Biomedical Engineering, Hainan University, Sanya, China, 2Key Laboratory of Biomedical Engineering of Hainan Province, Sanya, China
First Author:
Zhanbo Zhang
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
Co-Author(s):
Tianhao Zhou
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
Yong Yang
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
Chenfei Cao
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
Xiaodong Zhang
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
Pengcheng Li
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
Feng Yue
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
Junjie Zhuo
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
School of Biomedical Engineering, Hainan University|Key Laboratory of Biomedical Engineering of Hainan Province
Sanya, China|Sanya, China
Introduction:
Rabbits are crucial in neuroscience, yet their model development lags behind due to the absence of a high-res brain atlas. The existing MRI-based atlas (Muñoz-Moreno et al., 2013) is ex vivo and low-res, leading to discrepancies with in vivo brains (Bock et al., 2008). We've created a high-res in vivo rabbit brain template with SyGN (Avants et al., 2008, 2010) and provided segmentation maps for automated analysis.
Methods:
We used 10 adult male and 2 adult female New Zealand rabbits (provided by Hainan Medical University), weighing 2.2–2.5 kg. After 3 days of acclimatization, anesthesia was induced via intramuscular injection of Zoletil (Zoletil injection anesthetic, Virbac, France) at a dose of 0.3 ml/kg. After intubation was performed with the endotracheal tube placed above the root of the tongue, anesthesia was maintained using a ventilator and isoflurane (RWD, Shenzhen). MRI scans were conducted using a human knee coil on a GE Signa 5T MRI scanner.
The T1-weighted images were acquired using a Fast Spoiled Gradient Echo (FSPGR) sequence, with the following parameters: TR = 14.5 ms, TE = minimum, flip angle = 9°, readout FOV = 58 mm, phase FOV = 52 mm, readout resolution = 176, phase resolution = 100, slice thickness = 0.33 mm, and averages = 12. T2-weighted images were acquired using a Variable Flip-Angle 3D (Matrix) sequence, with the following parameters: TR = 3000 ms, TE = 342 ms, readout FOV = 52 mm, phase FOV = 52 mm, readout resolution = 160, phase resolution = 100, slice thickness = 0.33 mm, and averages = 12. The rabbits were positioned in a prone position with their bodies extended, and their heads were fixed at the center of the magnetic field using a custom acrylic head frame to ensure optimal signal-to-noise ratio. Prior to the main experiment, we conducted a preliminary study to determine the appropriate number of averages, comparing the signal-to-noise ratios for different averaging values.
We then used the Symmetric Group-Wise Normalization (SyGN) algorithm to generate the template image. The steps for the SyGN algorithm are as follows:
1. An initial target template is created by averaging the preprocessed images.
2. Each image is pairwise normalized (rigid transformation, affine transformation, and diffeomorphic registration) to the target template.
3. A new target template is created by averaging the normalized images from step 2.
4. The new target template is transformed using the average affine transformation and inverse deformation field, scaled by a factor of 0.25.
5. Steps 2 to 4 are iterated until convergence: the root mean square average difference between the continuous overall average template images output from step 4 no longer changes.
An unsupervised clustering method was used to classify the tissues into five categories: gray matter, white matter, cerebrospinal fluid, skull, and background, followed by manual adjustment of the segmentation results. The entire workflow is shown in Figure 1.
The T1-weighted images were acquired using a Fast Spoiled Gradient Echo (FSPGR) sequence, with the following parameters: TR = 14.5 ms, TE = minimum, flip angle = 9°, readout FOV = 58 mm, phase FOV = 52 mm, readout resolution = 176, phase resolution = 100, slice thickness = 0.33 mm, and averages = 12. T2-weighted images were acquired using a Variable Flip-Angle 3D (Matrix) sequence, with the following parameters: TR = 3000 ms, TE = 342 ms, readout FOV = 52 mm, phase FOV = 52 mm, readout resolution = 160, phase resolution = 100, slice thickness = 0.33 mm, and averages = 12. The rabbits were positioned in a prone position with their bodies extended, and their heads were fixed at the center of the magnetic field using a custom acrylic head frame to ensure optimal signal-to-noise ratio. Prior to the main experiment, we conducted a preliminary study to determine the appropriate number of averages, comparing the signal-to-noise ratios for different averaging values.
We then used the Symmetric Group-Wise Normalization (SyGN) algorithm to generate the template image. The steps for the SyGN algorithm are as follows:
1. An initial target template is created by averaging the preprocessed images.
2. Each image is pairwise normalized (rigid transformation, affine transformation, and diffeomorphic registration) to the target template.
3. A new target template is created by averaging the normalized images from step 2.
4. The new target template is transformed using the average affine transformation and inverse deformation field, scaled by a factor of 0.25.
5. Steps 2 to 4 are iterated until convergence: the root mean square average difference between the continuous overall average template images output from step 4 no longer changes.
An unsupervised clustering method was used to classify the tissues into five categories: gray matter, white matter, cerebrospinal fluid, skull, and background, followed by manual adjustment of the segmentation results. The entire workflow is shown in Figure 1.
·Data Acquisition and Processing Procedure ( a represents the data acquisition section, b represents the data preprocessing section, and c represents the template construction section )
Results:
We collected MRI brain images from 12 rabbits and used the SyGN algorithm to create T1-w and T2-w brain templates, both with a resolution of 0.33×0.33×0.33mm3. Additionally, corresponding tissue segmentations were obtained. The individual data, templates, tissue segmentation masks, and skull masks are shown in Figure 2.
·Collected individual data, templates, and tissue segmentation results
Conclusions:
We have developed a set of T1-weighted and T2-weighted brain templates, along with corresponding tissue segmentations. This is the first adult rabbit in vivo brain template with tissue segmentation, filling a significant gap in the field. By registering individual data to the template, we can calculate the volume of regions of interest at the voxel level. Additionally, the tissue segmentation maps can serve as priors for generating segmentations of individual data. In the future, we plan to make our data and templates publicly available to contribute to further research in the field.
Modeling and Analysis Methods:
Image Registration and Computational Anatomy 1
Segmentation and Parcellation 2
Keywords:
Data Registration
MRI
Other - Rabbit
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
Provide references using APA citation style.
Avants, B., Epstein, C., Grossman, M., & Gee, J. (2008). Symmetric diffeomorphic image registration with cross-correlation: Evaluating automated labeling of elderly and neurodegenerative brain. Medical Image Analysis, 12(1), 26–41.
Bock, N. A., Paiva, F. F., Nascimento, G. C., Newman, J. D., & Silva, A. C. (2008). Cerebrospinal fluid to brain transport of manganese in a non-human primate revealed by MRI. Brain Research, 1198, 160–170.
Muñoz-Moreno, E., Arbat-Plana, A., Batalle, D., Soria, G., Illa, M., Prats-Galino, A., Eixarch, E., & Gratacos, E. (2013). A Magnetic Resonance Image Based Atlas of the Rabbit Brain for Automatic Parcellation. PLoS ONE, 8(7), e67418.