Cognitive impairment in mild traumatic brain injury: a diffusion kurtosis imaging and volumetric study

Abstract

Background

A significant number of patients with mild traumatic brain injury (mTBI) would experience cognitive deficit.

Purpose

To investigate the brain structural changes in sub-acute mTBI by diffusion kurtosis imaging (DKI) and volumetric analysis, and to assess the relationship between brain structural changes and cognitive functions.

Material and Methods

A total of 23 patients with sub-acute mTBI and 24 control participants were recruited. All the participants underwent examinations of neuropsychological tests, DKI, and magnetic resonance imaging (MRI)-based morphological scans. Images were investigated using whole brain-based analysis and further regions of interest-based analysis for subcortical nuclei. The neuropsychological tests were compared between the mTBI and the control group. Correlation analysis was performed to examine the relationship between gray matter (GM) volume, DKI parameters, and cognitive functions.

Results

Compared with control participants, mTBI patients performed worse in the domains of verbal memory, attention and executive function (P < 0.05). No regional GM volume differences were observed between the mTBI and control groups (P > 0.05). Using DKI, patients with mTBI showed lower mean kurtosis (MK) in widespread white matter (WM) regions and several subcortical nuclei (P < 0.05), and higher mean diffusivity (MD) in the right pallidum (P < 0.05). Lower MK value of multiple WM regions and several subcortical nuclei correlated with cognitive impairment (P < 0.05).

Conclusion

DKI was sensitive in detecting brain microstructural changes in patients with sub-acute mTBI showing lower MK value in widespread WM regions and several subcortical nuclei, which were statistically associated with cognitive deficits.

Keywords

Mild traumatic brain injury diffusion kurtosis imaging voxel-based morphometry tract-based spatial statistics subcortical nuclei cognitive impairment

Introduction

Mild traumatic brain injury (mTBI) is a major public health problem, affecting approximately 42 million individuals worldwide, annually (1). Despite its name, mTBI could lead to various neurological and non-neurological disorders, among which cognitive impairment is the most common one (2). Previous studies revealed that a significant number of patients with mTBI experienced cognitive deficit during the first two weeks, and the cognitive deficit would persist for about three months after the injury (3,4). The possible mechanisms for cognitive impairment include both gray matter (GM) and/or white matter (WM) damage. Finding a reliable biomarker for accurate assessment of the brain pathological changes underlying cognitive deficit after mTBI is of great significance.

Magnetic resonance imaging (MRI), as a non-invasive tool, is playing an increasingly important role in the assessment of TBI from its acute to chronic phases (5). Notably, diffusion tensor imaging (DTI), which describes water diffusion in biological tissues, was frequently used to quantify the microstructural changes of WM (6). Previous studies have proved the value of DTI to assess WM microstructural changes in sub-acute mTBI using a region of interest (ROI) or tract-based spatial statistics (TBSS) method, and the DTI parameters changes in certain WM tracts were associated with cognitive impairment (7 –10).

However, DTI is insufficient to reflect the actual non-Gaussian diffusion of water molecules in brain tissues. In particular, it cannot accurately estimate the circumstance of WM with multiple fibers in different directions. Diffusion kurtosis imaging (DKI), in contrast, is a method that uses the non-Gaussian model of water diffusion that could provide complementary information such as mean kurtosis (MK) to depict the complexities of the brain microstructure, especially for GM (11). Theoretically, DKI is a more accurate method than DTI for investigating the pathological alterations in brain tissues. Presently, only a few studies have proved that DKI could provide more information regarding the mTBI-induced brain microstructural changes (12 –15).

In addition to WM tracts, GM damage was also observed in patients with mTBI. Loss of GM volume has been shown in patients with chronic mTBI by using a voxel-based morphometry (VBM) method (16 –18). However, whether the volume loss in mTBI begins to appear in an earlier sub-acute stage and its relationship with cognitive function is unknown. Furthermore, the study by Zagorchev et al. (19) observed volume loss in subcortical nuclei including hippocampus, amygdala, caudate, putamen, and thalamus in patients with chronic mTBI. As for the microstructural changes, only a few studies proved that DKI parameters in the thalamus might be useful for early prediction of persistent brain damage and cognitive outcome after mTBI (12,13,20). Other subcortical nuclei have not been studied using the DKI technique in cognitive impairment after mTBI.

Determining brain structural changes that underlie cognitive impairment will help reveal the pathophysiology of cognitive deficits due to mTBI. The aim of the present study was to combine DKI and volumetric analysis to assess the GM and WM microscopic structure changes in patients with sub-acute mTBI and, if present, to correlate these changes with the neuropsychological performance. We hypothesized that: (i) brain volume and DKI parameter alterations could be found in sub-acute mTBI and would be associated with cognitive impairment; and (ii) compared with DTI, DKI could detect subtle microstructural changes in sub-acute mTBI. These findings would enhance our knowledge regarding cognitive impairment due to mTBI.

Material and Methods

Participants

The present study was performed in accordance with the recommendations of the local Ethics Committee with written informed consent from all the participants in accordance with the Declaration of Helsinki. The study included 23 consecutive patients with mTBI in the sub-acute stage (from two weeks to two months) recruited from the neurosurgery clinic. mTBI was diagnosed based on the Mayo classification system for injury severity (21). In addition, 24 healthy participants without any history of mTBI were also included in this study. All the participants were assessed using a battery of neuropsychological tests followed by an MRI examination. The exclusion criteria were as follows: a history of drug abuse; cerebrovascular disease; psychiatric or neurological illness before the head injury; contraindications to MR scanning; and patients with obvious brain abnormality on conventional MRI.

Magnetic resonance imaging

The MRI scans were performed on a 3.0-T MRI scanner (Simens Verio, Erlangen, Germany) with a 32-channel head coil at our institution. A three-dimensional MPRAGE T1 sequence was used for measuring the GM volume with the following parameters: repetition time (TR)/echo time (TE)/inversion time = 1500/2.96/900 ms; flip angle = 9°; matrix = 256 × 256; field of view (FOV) = 230 mm; 120 axial slices; slice thickness = 1 mm. The DKI sequence was performed using a spin-echo echo-planar imaging diffusion sequence with 30 different diffusion-encoding directions. For each direction, six b-values (b = 0, 500, 1000, 1500, 2000, 2500 s/mm²) were acquired. Other imaging parameters were as follows: TR/TE = 5400/109 ms; matrix = 128 × 128; FOV = 230 mm; 35 axial slices; slice thickness = 3 mm. The scan duration for DKI was 14 min 4 s.

Image analysis

The whole-brain volumetric analysis was done using FSL-VBM, which is a voxel-based morphometry analysis method (22) and is a part of the FMRIB Software Library (FSL, version 5.0.1, Oxford, UK, http://www.fmrib.ox.ac.uk/fsl/) (23). As for the subcortical nuclei volumetric analysis, the “FIRST” toolbox (24) in the FSL software package was used to extract the bilateral volumes of the following six subcortical nuclei: hippocampus; thalamus; amygdala; caudate nucleus; putamen; and globus pallidus.

For DKI image analysis, motion artifacts and eddy-current–induced distortion were first corrected using the affine alignment of each diffusion-weighted image to the b0 image by employing the FSL software toolbox. Subsequently, the Diffusion Kurtosis Estimator software (DKE) (https://www.nitrc.org/projects/dke/) was used to calculate the DKI parametric maps including MK, as well as mean diffusivity (MD) and fractional anisotropy (FA) (25). The whole-brain DKI parameters were analyzed using TBSS (26) in the FSL. As for the subcortical nuclei DKI analysis, subcortical nuclei masks were aligned to the diffusion space using Flirt tool with nine degrees of freedom in FSL software package (27,28). All subcortical nuclei masks in the diffusion space were checked visually for registration errors and no errors were observed. All the parameters were manifested in the ITK-SNAP software (www.itksnap.org) (29). All the image analyses were done by an experienced board-certified neuroradiologist (M.L.W.). Supplemental Figure 1 depicts an example of subcortical nuclei segmentation and masks in the diffusion space.

Neuropsychological assessment

All participants completed a battery of neuropsychological tests that are sensitive to detect TBI-induced cognitive impairment on the same day, at which the participants underwent MRI. This study focused on three cognitive domains: verbal memory; attention; and executive function. Verbal memory was assessed using the Huashan Auditory Verbal Learning Test (HAVLT) (30). The total learning score for each participant was measured on the basis of the total score of three learning free-recall trials of 12 words. Furthermore, the attention abilities were examined using Digit Span Forward. The executive function was assessed using Digit Span Backward. Another experienced board-certified neuroradiologist (M.M.Y) performed the neuropsychological assessments.

Statistical analysis

Statistical tests were performed by the Statistical Package for Social Sciences (IBM Corp., Armonk, NY, USA) software for Windows version 20.0. The demographics, neuropsychological tests results, and subcortical nuclei MRI parameters between the mTBI group and the control group were compared using the chi-square test for qualitative variables and the Student’s t-test for normal distribution quantitative variables or the Mann–Whitney U-test for non-normal distribution quantitative variables. Pearson or Spearman correlation analysis was performed to examine the relationship between subcortical nuclei volume, DKI parameters and cognitive function. P values < 0.05 indicated statistical significance. The whole brain-based analysis was performed using a voxel-wise general linear model (GLM) for between-group comparisons and correlation analysis (age and gender included as covariates), with permutation-based non-parametric testing (5000 permutations). Threshold-free cluster enhancement (TFCE) (31) was used to form significant clusters, corrected for the multiple comparisons by family-wise error (FWE) rate. The corrected threshold for significance was P < 0.05.

Results

Patient demographics and neuropsychological assessment

Patient demographics and specific mTBI characteristics are shown in Table 1. The major reason for mTBI is traffic accident (15/23, 65.22%). Nineteen patients had a Glasgow Coma Scale (GCS) score of 15, three had a score of 14, and one had a score of 13. MRI and neuropsychological tests were performed at a mean of 39 days after the injury. Five patients with mTBI had cerebral microbleeds and no patients with mTBI were diagnosed with diffuse axonal injury. Compared with the control participants, the patients with mTBI had no significant differences in age, sex, and education. However, patients with mTBI performed worse in the domains of verbal memory, attention, and executive function (P < 0.05).

Table 1.

Patients demographics and neuropsychological results.

	Control	Mild TBI	P value*
Age (years)	49 ± 13.5	47 ± 9.9	0.551
Male gender^†	12 (50)	12 (52.17)	0.882
Education (years)	7.5 (6–12)	7.5 (6–9)	0.633
Time since injury (days)		39 ± 16.4
Memory performance
HAVLT (total)	15.92 ± 3.86	11.87 ± 5.49	0.010*
Attention
Digit span test (forward)^‡	8 (7.25–9)	8 (6–8)	0.029*
Executive function
Digit span test (backward)^‡	5 (4–5)	3 (3–4)	0.003*

Values are given as n (%), mean ± SD (using Student’s t-test), or median (IQR).

*P < 0.05.

^†Chi-square test.

^‡Mann–Whitney U-test.

HAVLT, Huashan Auditory Verbal Learning Test; IQR, interquartile range; SD, standard deviation; TBI, traumatic brain injury.

Whole brain-based analysis in brain DKI parameters and volume

In the whole brain-based volume analysis, no regional GM volume differences were observed between the mTBI and control groups (P > 0.05, FWE corrected). As for the whole brain-based DKI parameters analysis, patients with mTBI showed lower MK value in widespread WM regions (P < 0.05, FWE corrected) than control participants. Fig. 1 depicts the specific significant regions mainly including the intrahemispheric association fibers of bilateral superior longitudinal fasciculus (SLF), cingulum, right inferior longitudinal fasciculus (ILF), inferior fronto-occipital fasciculus (IFF) and uncinate fasciculus (UF), inter-hemispheric fibers of body of corpus callosum, projection fibers of corticospinal tract (CST), anterior thalamic radiation (ATR), and bilateral internal capsule and external capsule. However, no statistical significance of FA and MD was observed between the mTBI and control groups.

Fig. 1.

Axial images derived from TBSS results and rendered on T1-weighted MR images from the ch2bet atlas showing lower MK in widespread WM regions in the mTBI group than that in the control group (P < 0.05 FWE corrected for multiple comparisons). Red = decrease in patients with mTBI. FWE, family-wise error; MK, mean kurtosis; MR, magnetic resonance; mTBI, mild traumatic brain injury; TBSS, tract-based spatial statistics; WM, white matter.

Further correlational study revealed that attention deficit revealed by lower digit span forward score was significantly associated with lower MK value in multiple WM regions (P < 0.05, FWE corrected) in the mTBI group. Fig. 2 depicts the specific significant regions, mainly including: intrahemispheric association fibers of bilateral SLF, ILF, IFF, UF, and cingulum; inter-hemispheric fibers of body of corpus callosum; projection fibers of CST, ATR, bilateral internal capsule, external capsule, and middle cerebellar peduncle.

Fig. 2.

Axial images correlational analysis rendered on T1-weighted MR images from the ch2bet atlas showing positive correlations between MK values and digit span test forward scores in patients with mTBI (P < 0.05 FWE corrected for multiple comparisons). Red = positive correlation in patients with mTBI. FWE, family-wise error; MK, mean kurtosis; MR, magnetic resonance; mTBI, mild traumatic brain injury.

Abnormalities of DKI parameters and volume in subcortical nuclei

Similar to the whole brain-based volume analysis, no significant volume differences were observed in any subcortical nuclei between the mTBI and control groups (P > 0.05). Table 2 shows the DKI parameters of subcortical nulei in the mTBI group and the control group. The mTBI group had a lower MK value in the right hippocampus, left thalamus, left caudate, right putamen, and right pallidum than those in the control group (P < 0.05). The MD value of the right pallidum in the mTBI group was also higher than that in the control group (P < 0.05). In addition, we observed no statistical difference in the FA value of subcortical nuclei between the mTBI group and the control group.

Table 2.

Statistical analysis of subcortical nuclei DKI parameters between mTBI and control group.

	MK			MD
	Control	Mild TBI	P value	Control	Mild TBI	P value
Hippocampus, L	0.790 ± 0.097	0.747 ± 0.025	0.103	1.294 ± 0.115	1.295 ± 0.117	0.981
Hippocampus, R	0.805 ± 0.085	0.758 ± 0.028	0.047*	1.323 ± 0.075	1.307 ± 0.121	0.616
Thalamus, L	0.964 ± 0.127	0.891 ± 0.053	0.042*	1.167 ± 0.059	1.192 ± 0.101	0.344
Thalamus, R	0.966 ± 0.116	0.917 ± 0.068	0.135	1.228 ± 0.053	1.209 ± 0.110	0.491
Caudate, L	0.799 ± 0.112	0.728 ± 0.073	0.019*	1.342 ± 0.189	1.381 ± 0.212	0.596
Caudate, R	0.834 ± 0.115	0.786 ± 0.057	0.134	1.210 ± 0.153	1.212 ± 0.135	0.974
Putamen, L	0.969 ± 0.158	0.893 ± 0.068	0.086	0.813 ± 0.035	0.842 ± 0.055	0.105
Putamen, R	1.018 ± 0.155	0.932 ± 0.070	0.050*	0.814 ± 0.025	0.833 ± 0.054	0.257
Pallidum, L	1.310 ± 0.144	1.245 ± 0.108	0.109	0.793 ± 0.045	0.831 ± 0.113	0.276
Pallidum, R	1.335 ± 0.142	1.248 ± 0.104	0.031*	0.771 ± 0.045	0.830 ± 0.079	0.023*
Amygdala, L	0.792 ± 0.131	0.726 ± 0.041	0.065	1.102 ± 0.071	1.155 ± 0.101	0.115
Amygdala, R	0.792 ± 0.130	0.722 ± 0.045	0.054	1.087 ± 0.076	1.114 ± 0.113	0.746

Values are given as mean ± SD (Student’s t-test).

*P < 0.05.

L, left; MD, mean diffusivity; MK, mean kurtosis; R, right; TBI, traumatic brain injury.

Further correlational study revealed that the MK values of the left amygdala (r = 0.625, P = 0.010), right thalamus (r = 0.634, P = 0.008), right putamen (r = 0.582, P = 0.018), right hippocampus (r= 0.576, P = 0.020), and right amygdala (r = 0.503, P = 0.047) were positively correlated with the attention ability. The MD value of the left thalamus was negatively correlated with the verbal memory (r = –0.678, P = 0.022). The FA value of the left amygdala was positively correlated with the verbal memory (r = 0.611, P = 0.016) (Fig. 3).

Fig. 3

Correlational analysis between DKI parameters and neuropsychological scale scores. (a–e) Positive correlation between the MK of several subcortical nuclei with the attention ability revealed by the Digit Span Forward Score (P < 0.05). (f) Negative correlation between MD of the left thalamus with verbal memory revealed by Auditory Verbal Learning Test Score (P < 0.05). (g) Positive correlation between FA of the left amygdala with verbal memory revealed by Auditory Verbal Learning Test Score (P < 0.05). DKI, diffusion kurtosis imaging; FA, fractional anisotropy; MD, mean diffusivity; MK, mean kurtosis.

Discussion

The present study aimed to combine DKI and volumetric analysis to quantify the brain GM and WM microscopic structural changes in patients with sub-acute mTBI, and, if present, to correlate these changes with the neuropsychological test performance. Consistent with our hypothesis, DKI was sensitive to find brain microstructural changes revealing lower MK value in widespread WM regions and several subcortical nuclei in sub-acute mTBI, and were statistically associated with cognitive function changes. However, the whole brain-based analysis revealed no significant difference in the GM volume, FA, and MD value between the mTBI group and the control group.

Previous studies have revealed that the loss of GM volume was associated with cognitive impairment in patients with mTBI (16 –18). However, all the present studies mainly focused on chronic mTBI. Whether the loss of GM volume begins to appear in an earlier sub-acute stage is unknown. In our study, we observed no significant changes in GM volume including subcortical nuclei in patients with sub-acute mTBI. We speculated that the neuron loss might not be obvious in the sub-acute stage. To date, the “atrophy curve,” that is, the relationship between brain volume and injury time is still unknown (32). Our study indicates that the loss of GM volume would not appear at the sub-acute stage (from two weeks to two months) after mTBI.

As for the DKI parameters, our study found lower MK value in widespread WM regions and several subcortical nuclei in patients with sub-acute mTBI. The study by Grossman et al. (13) also revealed lower MK values in individuals with mTBI. However, the study by Davenport et al. (33) reported higher MK values in high school football players. In a recent study of individuals with sport-related concussion, the concussed group showed significantly higher axial kurtosis than the control group (34). The contradictory results are most likely caused by the difference in injury mechanism, as football may only cause subtle injury to the brain. Furthermore, there was no statistical significance in FA and MD between the mTBI group and control group in our study. As most of the individuals with mTBI (19/23) had a GCS of 15 in our study, the impact of mTBI on the brain could be relatively weak, and thus, the differences both in FA and MD would be insignificant. The study by Hart et al. (35) also found no statistical differences in DTI-derived FA and MD values using TBSS between control and university amateur boxers. Thus, the present study suggested the MK parameter was more sensitive than DTI parameters including FA and MD parameters in detecting brain microstructural changes after mTBI.

The underlying brain pathological changes of DKI parameters are not yet completely understood. TBI is a heterogeneous and complex pathological process (36), including the following: (i) cell death and inflammation; (ii) cell proliferation for tissue replacement; and (iii) tissue remodeling (37). Reasonably, the DKI parameter would reflect an integration of multiple events after the injury. Our previous animal study revealed that low MK value in the TBI rat brain was associated with neuron loss (38). The lower MK value might result at least partially from the contribution of degenerative processes like neuron loss and changes in axonal and myelin density (39), leading to the decrement of diffusional heterogeneity.

Furthermore, the present study showed lower MK values in multiple WM regions and several subcortical nuclei and correlated with the attention deficit. This was reasonable as the lower MK value in the mTBI group might represent neuron loss and demyelinating, which could further cause the disruption of the brain attention circuit including the basal ganglia, amygdala, SLF, and so on (40). Therefore, our study suggested the MK value could be used to assess cognitive impairment in mTBI.

The present study has some limitations. First, the sample size in the study is relatively small, and future studies with larger sample sizes are needed to replicate our results. Second, as the baseline volume and DKI data of subcortical nuclei before mTBI were not obtained, we cannot conclude that the DKI parameter differences were attributable to the injury. The DKI parameter differences might exist before mTBI occurred and even predispose an individual to have an mTBI. Third, this study was a cross-sectional study, and thus we cannot determine the exact order of causal effects between neuroimaging parameters changes and cognitive impairment. To determine their order of causal effect, future large longitudinal studies will be required to better assess the relationships between neuroimaging changes and cognitive impairment after mTBI. Fourth, the MD and FA values were derived from the DKI model rather than the DTI model. Comparison with previous studies using DTI-derived MD and FA values could not be made as this could cause confusion in interpretation.

In conclusion, the present study found that DKI was sensitive to detect brain microstructural changes in patients with sub-acute mTBI, showing lower MK value in widespread WM regions and several subcortical nuclei, which were statistically associated with cognitive deficits. No loss of GM volume was observed in patients with sub-acute mTBI.

Supplemental Material

sj-pdf-1-acr-10.1177_0284185121998317 - Supplemental material for Cognitive impairment in mild traumatic brain injury: a diffusion kurtosis imaging and volumetric study

Supplemental material, sj-pdf-1-acr-10.1177_0284185121998317 for Cognitive impairment in mild traumatic brain injury: a diffusion kurtosis imaging and volumetric study by Ming-Liang Wang, Xiao-Er Wei, Meng-Meng Yu and Wen-Bin Li in Acta Radiologica

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received the following financial support for the research, authorship, and/or publication of this article: This study is supported by National Natural Science Foundation of China (Nos. 81901727, 81271540, 81301213), Natural Science Foundation of Xinjiang Province (No. 2016D01C083), 2018 Youth Medical Talents – Medical Imaging Practitioner Program, and Shanghai key discipline of medical imaging (No. 2017ZZ02005).

ORCID iD

Wen-Bin Li

Supplemental Material

Supplemental material for this article is available online.

References

Gardner

Yaffe

Epidemiology of mild traumatic brain injury and neurodegenerative disease. Mol Cell Neurosci 2015; 66:75–80.

Masel

DeWitt

DS.

Traumatic brain injury: a disease process, not an event. J Neurotrauma 2010; 27:1529–1540.

Carroll

Cassidy

Cancelliere

, et al. Systematic review of the prognosis after mild traumatic brain injury in adults: cognitive, psychiatric, and mortality outcomes: results of the International Collaboration on Mild Traumatic Brain Injury Prognosis. Arch Phys Med Rehabil 2014; 95:S152–173.

Levin

Diaz-Arrastia

RR.

Diagnosis, prognosis, and clinical management of mild traumatic brain injury. Lancet Neurol 2015; 14:506–517.

Wang ML, Li WB. Cognitive impairment after traumatic brain injury: the role of MRI and possible pathological basis. J Neurol Sci 2016;370:244–250.

Basser

Mattiello

LeBihan

MR diffusion tensor spectroscopy and imaging. Biophys J 1994; 66:259–267.

Ryan

Genc

Beauchamp

, et al. White matter microstructure predicts longitudinal social cognitive outcomes after paediatric traumatic brain injury: a diffusion tensor imaging study. Psychol Med 2017; 48:679–691.

Strauss

Kim

Bidirectional changes in anisotropy are associated with outcomes in mild traumatic brain injury. AJNR Am J Neuroradiol 2016; 37:1983–1991.

Xiong

Zhu

Zhang

, et al. White matter integrity and cognition in mild traumatic brain injury following motor vehicle accident. Brain Res 2014; 1591:86–92.

10.

Wallace

Mathias

Ward

The relationship between diffusion tensor imaging findings and cognitive outcomes following adult traumatic brain injury: A meta-analysis. Neurosci Biobehav Rev 2018; 92:93–103.

11.

Jensen

Helpern

Ramani

, et al. Diffusional kurtosis imaging: the quantification of non-gaussian water diffusion by means of magnetic resonance imaging. Magn Reson Med 2005; 53:1432–1440.

12.

Grossman

Jensen

, et al. Thalamus and cognitive impairment in mild traumatic brain injury: a diffusional kurtosis imaging study. J Neurotrauma 2012; 29:2318–2327.

13.

Grossman

Jensen

Babb

, et al. Cognitive impairment in mild traumatic brain injury: a longitudinal diffusional kurtosis and perfusion imaging study. AJNR Am J Neuroradiol 2013; 34:951–957, s951–953.

14.

Lancaster

Meier

Olson

, et al. Chronic differences in white matter integrity following sport-related concussion as measured by diffusion MRI: 6-Month follow-up. Hum Brain Mapp 2018; 39:4276–4289.

15.

Karlsen

Einarsen

Moe

, et al. Diffusion kurtosis imaging in mild traumatic brain injury and postconcussional syndrome. J Neurosci Res 2019; 97:568–581.

16.

Zhou

Kierans

Kenul

, et al. Mild traumatic brain injury: longitudinal regional brain volume changes. Radiology 2013; 267:880–890.

17.

Little

Geary

Moynihan

, et al. Imaging chronic traumatic brain injury as a risk factor for neurodegeneration. Alzheimers Dement 2014; 10:S188–195.

18.

Patel JB, Wilson SH, Oakes TR, et al. Structural and volumetric brain MRI findings in mild traumatic brain injury. Am J Neuroradiol 2020;41:92–99.

19.

Zagorchev

Meyer

Stehle

, et al. Differences in regional brain volumes two months and one year after mild traumatic brain injury. J Neurotrauma 2016; 33:29–34.

20.

Stokum

Sours

Zhuo

, et al. A longitudinal evaluation of diffusion kurtosis imaging in patients with mild traumatic brain injury. Brain Inj 2015; 29:47–57.

21.

Malec

Brown

Leibson

, et al. The mayo classification system for traumatic brain injury severity. J Neurotrauma 2007; 24:1417–1424.

22.

Ashburner

Friston

KJ.

Voxel-based morphometry–the methods. Neuroimage 2000; 11:805–821.

23.

Smith

Jenkinson

Woolrich

, et al. Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage 2004; 23(Suppl. 1):S208–219.

24.

Patenaude

Smith

Kennedy

, et al. A Bayesian model of shape and appearance for subcortical brain segmentation. Neuroimage 2011; 56:907–922.

25.

Tabesh

Jensen

Ardekani

, et al. Estimation of tensors and tensor-derived measures in diffusional kurtosis imaging. Magn Reson Med 2011; 65:823–836.

26.

Smith

Jenkinson

Johansen-Berg

, et al. Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data. Neuroimage 2006; 31:1487–1505.

27.

Jenkinson

Smith

A global optimisation method for robust affine registration of brain images. Med Image Anal 2001; 5:143–156.

28.

Jenkinson

Bannister

Brady

, et al. Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage 2002; 17:825–841.

29.

Yushkevich

Piven

Hazlett

, et al. User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage 2006; 31:1116–1128.

30.

Duan

Zheng

Cognitive impairments associated with periventricular white matter hyperintensities are mediated by cortical atrophy. Acta Neurol Scand 2014; 130:178–187.

31.

Smith

Nichols

TE.

Threshold-free cluster enhancement: addressing problems of smoothing, threshold dependence and localisation in cluster inference. Neuromage 2009; 44:83–98.

32.

Ross

DE.

Review of longitudinal studies of MRI brain volumetry in patients with traumatic brain injury. Brain Inj 2011; 25:1271–1278.

33.

Davenport

Apkarian

Whitlow

, et al. Abnormalities in diffusional kurtosis metrics related to head impact exposure in a season of high school varsity football. J Neurotrauma 2016; 33:2133–2146.

34.

Muftuler

Meier

Keith

, et al. Serial diffusion kurtosis magnetic resonance imaging study during acute, subacute, and recovery periods after sport-related concussion. J Neurotrauma 2020; 37:2081–2092.

35.

Hart

Housden

Suckling

, et al. Advanced magnetic resonance imaging and neuropsychological assessment for detecting brain injury in a prospective cohort of university amateur boxers. Neuroimage Clin 2017; 15:194–199.

36.

Kou

VandeVord

PJ.

Traumatic white matter injury and glial activation: from basic science to clinics. Glia 2014; 62:1831–1855.

37.

Burda

Sofroniew

MV.

Reactive gliosis and the multicellular response to CNS damage and disease. Neuron 2014; 81:229–248.

38.

Wang M-L, Yu M-M, Yang D-X, et al. Diffusion kurtosis imaging characterizes brain microstructural changes associated with cognitive impairment in a rat model of chronic traumatic brain injury. Neuroscience 2018;392:180–189.

39.

Fieremans

Jensen

Helpern

JA.

White matter characterization with diffusional kurtosis imaging. Neuroimage 2011; 58:177–188.

40.

Mueller

Hong

Shepard

, et al. Linking ADHD to the neural circuitry of attention. Trends Cogn Sci 2017; 21:474–488.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.35 MB