Machine learning based multimodal MRI analysis of brain changes after repetitive head impacts
Poster No:
1976
Submission Type:
Abstract Submission
Authors:
Katze Zambo1, Eryn Kwon1,2, Joshua McGeown2, Christian John Saludar1, Justin Fernandez1,2, Alan Wang1,3, Samantha Holdsworth2,3, Vickie Shim1
Institutions:
1Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand, 2Mātai Medical Research Institute, Gisborne, New Zealand, 3Faculty of Medical and Health Sciences & Centre for Brain Research, University of Auckland, Auckland, New Zealand
First Author:
Co-Author(s):
Eryn Kwon
Auckland Bioengineering Institute, University of Auckland|Mātai Medical Research Institute
Auckland, New Zealand|Gisborne, New Zealand
Auckland Bioengineering Institute, University of Auckland|Mātai Medical Research Institute
Auckland, New Zealand|Gisborne, New Zealand
Christian John Saludar
Auckland Bioengineering Institute, University of Auckland
Auckland, New Zealand
Auckland Bioengineering Institute, University of Auckland
Auckland, New Zealand
Justin Fernandez
Auckland Bioengineering Institute, University of Auckland|Mātai Medical Research Institute
Auckland, New Zealand|Gisborne, New Zealand
Auckland Bioengineering Institute, University of Auckland|Mātai Medical Research Institute
Auckland, New Zealand|Gisborne, New Zealand
Alan Wang
Auckland Bioengineering Institute, University of Auckland|Faculty of Medical and Health Sciences & Centre for Brain Research, University of Auckland
Auckland, New Zealand|Auckland, New Zealand
Auckland Bioengineering Institute, University of Auckland|Faculty of Medical and Health Sciences & Centre for Brain Research, University of Auckland
Auckland, New Zealand|Auckland, New Zealand
Samantha Holdsworth
Mātai Medical Research Institute|Faculty of Medical and Health Sciences & Centre for Brain Research, University of Auckland
Gisborne, New Zealand|Auckland, New Zealand
Mātai Medical Research Institute|Faculty of Medical and Health Sciences & Centre for Brain Research, University of Auckland
Gisborne, New Zealand|Auckland, New Zealand
Introduction:
Mild traumatic brain injury (mTBI) is a common injury worldwide, typically occurring from contact sports or accidents (Ropper, 2007). Repeated mTBIs can lead to neurodegenerative disorders such as chronic traumatic encephalopathy (McKee, 2015). Additionally, there is evidence that even subconcussive impacts, not meeting the threshold for a mTBI induce structural and functional changes (Bailes, 2013). Despite their prevalence, diagnosis of mTBIs is largely limited to neurocognitive testing, rather than objective biomarkers (Bodien, 2021), and the etiology of mTBIs is still not fully understood.
Advanced MRI is a powerful tool for investigating brain changes before and after mTBIs. At the time of injury, multiple tissue types - cerebral vessels, white matter, and grey matter - are affected. This study aims to analyse multimodal MRI datasets to examine the effects of head impacts on these tissues, both collectively and individually, to better understand the multifaceted impact of mTBIs on brain anatomy and physiology (Dean, 2014), with the ultimate goal of improving diagnostic methods.
Advanced MRI is a powerful tool for investigating brain changes before and after mTBIs. At the time of injury, multiple tissue types - cerebral vessels, white matter, and grey matter - are affected. This study aims to analyse multimodal MRI datasets to examine the effects of head impacts on these tissues, both collectively and individually, to better understand the multifaceted impact of mTBIs on brain anatomy and physiology (Dean, 2014), with the ultimate goal of improving diagnostic methods.
Methods:
We used data from a longitudinal study on the effects of subconcussive impacts on teenage rugby union athletes in Gisborne, New Zealand. 36 subjects were initially in the study, with 11 subjects completing all sessions during data collection (fig 1) (Tayebi, 2022). Subjects were scanned before, during, and after the season, using multiple MRI modalities including resting-state fMRI and 4DFlow. To the best of our knowledge, 4DFlow has not been used in mTBI studies.
Cerebral blood flow was analysed using QVTPlus (Dempsey, 2024), a quantitative tool for analysing 4DFlow data. Volumetric flow rates were then measured from the left and right internal carotids and the basilar artery.
Resting-state fMRI images were processed using an in-house automated pipeline, which included motion correction, segmentation, and coregistration to the subject's T1-weighted anatomical images.
Changes in the DMN were compared using correlation of each session's DMN activation to the MNI152 atlas, summing the voxels in the correlation map.
We measured head impacts players experienced during practices and games using instrumented mouthguards (iMG), which recorded linear and angular accelerations of the head impact events. Symptom data was also collected using the Sports Concussion Assessment Tool 5 (SCAT5) (Echemendia, 2017). We divided our cohort into low and high impact groups, with 7 and 4 subjects respectively, via their exposure to cumulative angular acceleration throughout the rugby season (fig 1). Principal component analysis (PCA) was used to compare separation of our low and high impact groups into distinct groups.
Cerebral blood flow was analysed using QVTPlus (Dempsey, 2024), a quantitative tool for analysing 4DFlow data. Volumetric flow rates were then measured from the left and right internal carotids and the basilar artery.
Resting-state fMRI images were processed using an in-house automated pipeline, which included motion correction, segmentation, and coregistration to the subject's T1-weighted anatomical images.
Changes in the DMN were compared using correlation of each session's DMN activation to the MNI152 atlas, summing the voxels in the correlation map.
We measured head impacts players experienced during practices and games using instrumented mouthguards (iMG), which recorded linear and angular accelerations of the head impact events. Symptom data was also collected using the Sports Concussion Assessment Tool 5 (SCAT5) (Echemendia, 2017). We divided our cohort into low and high impact groups, with 7 and 4 subjects respectively, via their exposure to cumulative angular acceleration throughout the rugby season (fig 1). Principal component analysis (PCA) was used to compare separation of our low and high impact groups into distinct groups.
Results:
PCA was first performed with 4DFlow and fMRI separately. In both cases, the PCA input included the midseason-to-postseason cumulative linear acceleration and cumulative angular acceleration. The Mahalanobis distance between the low and high impact groups was 1.77 and 1.72 for fMRI and 4DFlow, respectively (fig 2a, fig 2b).
When 4DFlow, fMRI, and mouthguard data were combined, the distance between groups was reduced to 1.68, and with the addition of SCAT5 it was 1.65 (fig 2c, fig 2d).
When 4DFlow, fMRI, and mouthguard data were combined, the distance between groups was reduced to 1.68, and with the addition of SCAT5 it was 1.65 (fig 2c, fig 2d).
Conclusions:
The strong separation in the 4DFlow and mouthguard PCA indicates its applicability to mTBI study. The addition of SCAT5 data weakened the separation, but slightly. Multimodal PCA produced weaker separation than either unimodal PCA, and more so when SCAT5 was introduced, but only minimally. Our results indicate that advanced MRI can be a potential imaging biomarker for detecting brain changes after repetitive subconcussive impacts. However, the combined effects of 4DFlow and fMRI might require other data-driven methods that account for the dynamic nature of these signals.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 2
Modeling and Analysis Methods:
fMRI Connectivity and Network Modeling
Methods Development
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Neuroanatomy Other
Novel Imaging Acquisition Methods:
Multi-Modal Imaging 1
Keywords:
Cerebral Blood Flow
FUNCTIONAL MRI
Machine Learning
Other - Mild Traumatic Brain Injury
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:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
Bodien, Y. G. (2021). Diagnosing level of consciousness: The limits of the Glasgow Coma Scale total score. Journal of Neurotrauma, 38(23), 3295–3305.
Dean, P. J. A. (2014). Multimodal imaging of mild traumatic brain injury and persistent postconcussion syndrome. Brain and Behavior, 5(1), e00292.
Dempsey, S. (2024). Measuring global cerebrovascular pulsatility transmission using 4D flow MRI. Scientific Reports, 14(1), 12604.
Echemendia, R. J. (2017). The sport concussion assessment tool 5th edition (SCAT5). British Journal of Sports Medicine, 51(11), 848–850.
McKee, A. C. (2015). The neuropathology of chronic traumatic encephalopathy. Brain Pathology, 25(3), 350–364.
Ropper, A. H. (2007). Concussion. New England Journal of Medicine, 356(2), 166–172.
Tayebi, M. (2022). What happens to the brain over a single season of playing high school rugby: Structural and white matter fibre tract changes related to impact. In Proceedings of the International Society for Magnetic Resonance in Medicine, 30, 3045.