0118
A pediatric brain template for myelin water fraction and diffusion tensor imaging.1Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada, 2International Collaboration on Repair Discoveries, Vancouver, BC, Canada, 3Radiology, University of British Columbia, Vancouver, BC, Canada, 4Educational and Counseling Psychology, and Special Education, University of British Columbia, Vancouver, BC, Canada, 5Phillips Healthcare, Hamburg, Germany, 6Medicine, University of British Columbia, Vancouver, BC, Canada, 7Biomedical Engineering, University of Alberta, Edmonton, AB, Canada, 8Pathology and Laboratory Medicine, University of British Columbia, Vancouver, BC, Canada
Synopsis
Measuring pediatric brain myelination can provide insights into normal brain development as well as many pediatric brain disorders. We have created a myelin water and diffusion tensor template for healthy children aged 9-10 years to be used as a reference in imaging studies. Our template was produced using ANTs software to provide high-quality anatomical alignment of brain structures. ROI analysis revealed significant differences in myelin water fraction between children and adults. We found no significant correlation between myelin water fraction and diffusion tensor metrics across the ROIs investigated, highlighting the complementary information these two techniques provide.
Introduction
Accurate measurement of myelination is important for the investigation of normal childhood cognitive development and many pediatric brain disorders. Several advanced imaging methods may be suitable for assessing myelin in the developing brain. Myelin water imaging (MWI) measures multi-exponential T2 relaxation to isolate the short-T2 water trapped between myelin bilayers (myelin water fraction (MWF), a histopathology-validated marker of myelin density1). Diffusion tensor imaging (DTI) can also be used to measure white matter structural development. The fractional anisotropy (FA) of water diffusion is affected by many elements of white matter microstructure including axon density, diameter, alignment and myelination, many of which increase during brain maturation. FA, along with other diffusion metrics from diffusion tensor analysis such as mean diffusivity (MD), axial diffusivity (AD) and radial diffusivity (RD) have been investigated as neural correlates for cognitive development2–4.
When analysing pediatric MWI and DTI data, a high-quality study-specific template is necessary, as registration to an adult standard space can cause increased deformation and bias5. Age-specific control data for reference is also needed, as myelination increases rapidly during childhood. We have created a pediatric structural, diffusion and MWF template for healthy children aged 9-10yrs to use for registration and comparison in pediatric studies. We hope to extend this template to other ages in the future.
Methods
Acquisition: Twenty healthy children (mean age 9.6±0.3yrs, 4 female) in Grade 4 were imaged using a 3T Philips Achieva MRI scanner with an 8 channel SENSE head-coil to acquire: (1) 3DT1-weighted structural images (IR-TFE, TR/TE=9.816/5.70ms, shot interval=3000ms, TI=876.6ms, reconstructed resolution=1.1x1.1x1.1mm); (2) MWI (48-echo GRASE, TR/TE=1000ms/8ms, ∆TE=8ms, acquisition resolution=1.5x1.8x4.4mm, reconstructed resolution=1.1x1.1x1.1mm, 10 slices) and (3) DTI (diffusion-weighted SE, TR/TE=3600/68ms, b-values=0 and 1000 s/mm2, 15 diffusion directions, acquisition resolution=2.2x2.2x2.2mm, reconstructed resolution=1.1x1.1x1.1mm).
Analysis: MWI data was fit with a multi-exponential decay using a non-negative least squares algorithm with spatial regularisation and stimulated echo correction for T2 components logarithmically spaced between 10ms and 2s6. Maps of MWF (defined as the signal fraction with T2 < 40ms) were produced. DTI data was motion and eddy-current corrected and fit with FSL’s ‘eddy’ and ‘dtifit’ algorithms to produce FA, MD, RD and AD maps 7,8.
Template creation: The pediatric template was created using Advanced Normalisation Tools (ANTs)9,10 which outperforms cross-sectional template analysis methods, especially with pediatric brain data11–13. Structural images were corrected for low-frequency bias and normalized to their mean intensity before being combined to create a representative template14. Template creation used a combination of rigid, affine, and symmetric diffeomorphic normalisation transformation algorithms. The quantitative maps for each subject were linearly registered to the T1-weighted scans and then transformed to the template space using ANTs. Finally, voxel-wise metric mean and standard deviation maps were calculated from all twenty subjects (Figure 1). The standard deviation in all metric maps was low throughout the brain and without clear anatomical structure (as seen in the representative slices in Figure 2) indicating that anatomical alignment was good.
Seven regions of interest (ROIs) were chosen from the JHU Tractography atlas15–17 and registered to the metric maps using FSL ‘fnirt’ (Figure 3). Mean and standard deviation of MWF, FA, MD, RD and AD across the 20 subjects was calculated from the voxels in each ROI (MWF and FA are shown in Figure 3a,b). Two-tailed unpaired t-tests were performed to compare pediatric MWF with adult values from a previous 3T study18 and also to compare metric values between the sexes in each ROI. Spearman correlations were examined between MWF and DTI myelin-related metrics (FA, RD).
Results and Discussion
Rapid myelination has been demonstrated in early childhood using the mcDESPOT MWI acquisition19. We found that pediatric MWF values in all of the ROIs were significantly lower than adult MWF (Figure 4), which was expected as cognitive development and brain maturation continue throughout the teenage years. No metric was significantly different between sexes (Figure 5), however this may be due to the small number of female participants in our study. FA and RD, the two measures thought to be most closely related to myelination, were not correlated with MWF (Figure 3c,d), although RD displayed a clear trend (p=0.11).Conclusion
We have created myelin and diffusion templates for children aged 9-10yrs which can be used as a reference in pediatric brain imaging studies. We found no significant relationships between MWF and diffusion metrics, suggesting these two methods are sensitive to different aspects of neurodevelopment and provide complementary information. All of the pediatric ROIs had significantly lower MWF than adult values, highlighting the importance of age-specific control data.Acknowledgements
We thank the study participants and the excellent MRI technologists at the UBC MRI Research Centre. This study was funded by the Alberta Heritage Foundation for Medical Research, NSERC, CIHR and the Canadian Language & Literacy Research Network.References
1. Laule, C. et al. Myelin water imaging in multiple sclerosis: quantitative correlations with histopathology. Mult. Scler. Houndmills Basingstoke Engl. 2006; 12, 747–753.
2. Snook, L., Paulson, L.-A., Roy, D., Phillips, L. & Beaulieu, C. Diffusion tensor imaging of neurodevelopment in children and young adults. NeuroImage. 2005; 26, 1164–1173.
3. Moseley, M. Diffusion tensor imaging and aging – a review. NMR Biomed. 2002; 15, 553–560.
4. Nagy, Z., Westerberg, H. & Klingberg, T. Maturation of White Matter is Associated with the Development of Cognitive Functions during Childhood. J. Cogn. Neurosci. 2004; 16, 1227–1233.
5. Yoon, U., Fonov, V. S., Perusse, D. & Evans, A. C. The effect of template choice on morphometric analysis of pediatric brain data. NeuroImage. 2009; 45, 769–777.
6. Whittall K.P., MacKay A.L. et al. In vivo measurement of T2 distributions and water contents in normal human brain. Magn. Reson. Med. 1997; 37: 34−43.
7. Behrens, T. E. J. et al. Characterization and propagation of uncertainty in diffusion-weighted MR imaging. Magn. Reson. Med. 2003; 50, 1077–1088.
8. Behrens, T. E. J., Berg, H. J., Jbabdi, S., Rushworth, M. F. S. & Woolrich, M. W. Probabilistic diffusion tractography with multiple fibre orientations: What can we gain? NeuroImage. 2007; 34, 144–155.
9. Avants, B. B., Epstein, C. L., Grossman, M. & Gee, J. C. Symmetric diffeomorphic image registration with cross-correlation: evaluating automated labeling of elderly and neurodegenerative brain. Med. Image Anal. 2008; 12, 26–41.
10. Avants, B. B. et al. A Reproducible Evaluation of ANTs Similarity Metric Performance in Brain Image Registration. NeuroImage. 2011; 54, 2033–2044.
11. Lawson, G. M., Duda, J. T., Avants, B. B., Wu, J. & Farah, M. J. Associations between children’s socioeconomic status and prefrontal cortical thickness. Dev. Sci. 2013; 16, 641–652.
12. Avants, B. B. et al. The optimal template effect in hippocampus studies of diseased populations. NeuroImage. 2010; 49, 2457–2466.
13. Avants, B. B. et al. The pediatric template of brain perfusion. Sci. Data. 2015; 2, 150003.
14. Tustison, N. J. et al. N4ITK: improved N3 bias correction. IEEE Trans. Med. Imaging. 2010; 29, 1310–1320.
15. Mori, S., Wakana, S., Zijl, P. C. M. van & Nagae-Poetscher, L. M. MRI Atlas of Human White Matter. (Elsevier, 2005).
16. Wakana, S. et al. Reproducibility of quantitative tractography methods applied to cerebral white matter. NeuroImage. 2007; 36, 630–644.
17. Hua, K. et al. Tract Probability Maps in Stereotaxic Spaces: Analyses of White Matter Anatomy and Tract-Specific Quantification. NeuroImage. 2008; 39, 336–347.
18. Meyers, S. M. et al. Reproducibility of myelin water fraction analysis: a comparison of region of interest and voxel-based analysis methods. Magn. Reson. Imaging. 2009; 27, 1096–1103.
19. Deoni, S. C. L. et al. Mapping Infant Brain Myelination with Magnetic Resonance Imaging. J. Neurosci. 31, 2011; 784–791.
Figures
