Studying new axial diffusion properties at high diffusion-weighting for the at-risk aging brain
Poster No:
1597
Submission Type:
Abstract Submission
Authors:
Carson Ingo1, Thomas Barrick2, Farzaneh Sorond1
Institutions:
1Northwestern University, Chicago, IL, 2St. George's, University of London, London, United Kingdom
First Author:
Co-Author(s):
Introduction:
Diffusion MRI (dMRI) is an emerging modality to study aging in the brain aside from classical FLAIR for white matter hyperintensities (WMH) (1). In this study, we utilize a generalized mathematical expression for diffusion signal decay, the Mittag-Leffler function (MLF), with α as an index to encode heterogeneous, power-law behavior, 0<α≤1 (2). As α decreases, diffusion becomes more heterogeneous, and is indicative of an increasingly complex diffusion environment (3). Here, we use tract-based spatial statistics (TBSS) (4) to study the sensitivity of α to white matter microstructure in an aging cohort that is at risk of stroke (5). We also evaluated the behavior of α at ultra-high b-values in an excised mouse brain.
Methods:
72 participants (57.0(3.4) years, 35 female) had vascular risk information collected since 1985 (5). Within this cohort, an at-risk sub-group was identified (57.1(3.42) years, 26 participants, 14 female) with significantly increased systolic blood pressure exposure of 124.3(16.3) mmHg compared to the control sub-group of 112.4(13.2) mmHg (p=0.02). dMRI data were collected on a 3T Siemens Prisma scanner for all participants with the following using SE-EPI: TE/TR=76.8/3000ms, resolution=1.5x1.5x1.5mm3, b-values=0,1000,2000,3000s/mm2, b=0s/mm2 averages=9, gradient directions=60. FLAIR data were acquired with the following: TE/TR/TI=289/6000/2200ms, flip=120°, resolution=1x1x1mm3 in order to identify visible white matter hyperintensities (WMH).
Open source dMRI data of a mouse brain was made available (6) and was acquired on a 9.4T Bruker Biospec scanner with the following using a PGSE-EPI: TE/TR=36.8/4000ms, resolution=120x120x400um3, b-values=0,1000,2500,4000,5500,7000,8500,10000,12500s/mm2, gradient directions=40.
All dMRI data were preprocessed as previously described (7). For both human and mouse data, the classical DTI parameters were calculated from 0 and 1000s/mm2 b-value shells to generate FA, MD, RD, and AD maps. α was estimated using all b-value shells the MLF (7,8). Akin to classical DTI which computes the tensor of the apparent diffusion coefficient (D), the tensor elements of α can also be computed with an analogous form as Fractional Anisotropy of α (FAα), Mean α (Mα), Radial α (Rα), and Axial α (Aα) (7,8).
TBSS was performed using the DTI and α metrics to identify for potential sensitivity to the sub-group with increased vascular risk versus the control sub-group not at-risk. The t-tests were adjusted for age, sex, and WMH volumes as covariates.
Open source dMRI data of a mouse brain was made available (6) and was acquired on a 9.4T Bruker Biospec scanner with the following using a PGSE-EPI: TE/TR=36.8/4000ms, resolution=120x120x400um3, b-values=0,1000,2500,4000,5500,7000,8500,10000,12500s/mm2, gradient directions=40.
All dMRI data were preprocessed as previously described (7). For both human and mouse data, the classical DTI parameters were calculated from 0 and 1000s/mm2 b-value shells to generate FA, MD, RD, and AD maps. α was estimated using all b-value shells the MLF (7,8). Akin to classical DTI which computes the tensor of the apparent diffusion coefficient (D), the tensor elements of α can also be computed with an analogous form as Fractional Anisotropy of α (FAα), Mean α (Mα), Radial α (Rα), and Axial α (Aα) (7,8).
TBSS was performed using the DTI and α metrics to identify for potential sensitivity to the sub-group with increased vascular risk versus the control sub-group not at-risk. The t-tests were adjusted for age, sex, and WMH volumes as covariates.
Results:
Fig.1 shows significantly decreased FAα, Mα, and Aα for the at-risk sub-group compared to the control sub-group not at-risk, however Rα did not reach significance. None of the classical DTI metrics (FA, MD, RD, AD) reached significance in the same group comparison. Decreased FAα and Mα were primarily driven by widespread decreased Aα for the at-risk sub-group, which demonstrated the greatest number of significantly different voxels out of all tested parameters (71.5% of TBSS skeleton voxels).
Fig.2a shows a coronal slice through an excised mouse brain for FA and FAα and magnification in the corpus callosum of their associated primary eigenvector orientations with a 9.48(0.89)° vector difference, which demonstrates good agreement with respect to orientation along the primary fiber direction. Fig.2b shows the dMRI signal fits in each direction along all ultra-high b-values demonstrating high quality fitting with low residual sum of square error of 0.29(0.03).
Fig.2a shows a coronal slice through an excised mouse brain for FA and FAα and magnification in the corpus callosum of their associated primary eigenvector orientations with a 9.48(0.89)° vector difference, which demonstrates good agreement with respect to orientation along the primary fiber direction. Fig.2b shows the dMRI signal fits in each direction along all ultra-high b-values demonstrating high quality fitting with low residual sum of square error of 0.29(0.03).
·Fig.1: At-Risk Sub-Group vs Control Sub-Group. Fractional Anisotropy of α (FAα), Mean α (Mα), and Axial α (Aα) are significantly decreased in the At-Risk Sub-Group (Red). Significance was p<0.05.
·Fig.2: A) Classical Fractional Anisotropy (FA), Fractional Anisotropy of α (FAα), and primary eigenvector alignment B) Example dMRI fit of the MLF in 40 directions in b-values up to 12,500 s/mm^2.
Conclusions:
Prior work is supportive of increased cellular heterogeneity in the presence of disease (9). In this study of white matter using human and mouse dMRI data, our results suggest that Aα and its well aligned eigenvector may be early biomarkers of axonal disruption, increased inflammation, and increased glial proliferation parallel to the primary fiber direction in white matter, prior to changes in classical DTI metrics (10).
Lifespan Development:
Aging 2
Modeling and Analysis Methods:
Diffusion MRI Modeling and Analysis 1
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Anatomy and Functional Systems
White Matter Anatomy, Fiber Pathways and Connectivity
Novel Imaging Acquisition Methods:
Diffusion MRI
Keywords:
ADULTS
Aging
HIGH FIELD MR
MRI
Neurological
NORMAL HUMAN
White Matter
WHITE MATTER IMAGING - DTI, HARDI, DSI, ETC
1|2Indicates the priority used for review
Provide references using author date format
2. Mittag-Leffler G. (1905). Sur la representation analytique d'une branche uniforme d'une fonction monogene. Acta Mathematica. 29(1):101-181.
3. Metzler R, Klafter J. (2000). The random walk's guide to anomalous diffusion: a fractional dynamics approach. Phys Rep. 339(1):1-77. doi: 10.1016/S0370-1573(00)00070-3.
4. Smith SM, Jenkinson M, Johansen-Berg H, et al. (2006). Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data. NeuroImage. 31(4):1487-505. doi:10.1016/j.neuroimage.2006.02.024.
5. Hughes GH, Cutter G, Donahue R, et al. (1987). Recruitment in the Coronary Artery Disease Risk Development in Young Adults (Cardia) Study. Control Clin Trials. 8(4 Suppl):68S-73S. doi:10.1016/0197-2456(87)90008-0.
6. Ianuş, A., Carvalho, J., Fernandes, F. F., Cruz, R., Chavarrias, C., Palombo, M., & Shemesh, N. (2022). Soma and Neurite Density MRI (SANDI) of the in-vivo mouse brain and comparison with the Allen Brain Atlas. Neuroimage. 254, 119135.
7. Barrick TR, Spilling CA, Ingo C, et al. (2020). Quasi-diffusion magnetic resonance imaging (QDI): A fast, high b-value diffusion imaging technique. NeuroImage. 211:116606. doi:10.1016/j.neuroimage.2020.116606.
8. Spilling CA, Howe FA, Barrick TR. (2022). Optimization of quasi-diffusion magnetic resonance imaging for quantitative accuracy and time-efficient acquisition. Magn Reson Med. 88(6):2532-2547. doi:10.1002/mrm.29420.
9. Gatto RG, Ye AQ, Colon-Perez L, et al. (2019). Detection of axonal degeneration in a mouse model of Huntington's disease: comparison between diffusion tensor imaging and anomalous diffusion metrics. Magn Reson Mater Phy. 32(4):461-471.
10. Walhovd KB, Johansen-Berg H, Karadottir RT. (2014). Unraveling the secrets of white matter--bridging the gap between cellular, animal and human imaging studies. Neuroscience. 276:2-13. doi:10.1016/j.neuroscience.2014.06.058.