Yi-Tien Li

Taipei Medical University | TMU · Neuroscience Research Ceter

Mild traumatic brain injury (mTBI) may cause significant microstructural changes of the cerebral white matter by the impact-accelerating shear force which may lead to the impairment of brain function, memory impairment, and even depression in severe cases. These white matter injuries cannot be visible by routine computed tomography or magnetic resonance imaging (MRI), hence, a reliable biomarker for clinical application is lacking. Most studies compared the group average difference among healthy control and mTBI, rendering little information on who might suffer from the prolonged neurocognitive symptoms or what might have caused the structural or functional underpinnings of disease progression. Quantitative susceptibility mapping (QSM) and myelin water fraction (MWF) are the advanced quantitative MRI techniques that enables us to detect myelin distribution and density for the patients with mTBI. Quantitative values derived from these quantitative MRI parametric maps are desirable as the true- quantification and vendor-independent measures which might be helpful for longitudinal evaluation of mTBI patients in the individual level. Genetic factors may also modulate the axonal and/or myelin degeneration and repairment as well as the vulnerability to secondary injury after mTBI, thus may account for some of the unexplained variation in the individual level. Identifying the relationship between genetic signatures, imaging features, and outcome endophenotypes may produce the opportunities for early intervention, risk stratification and prognostication for mTBI. In this 3-year project, we thus aim to conduct a translational research of white matter myelin degeneration and repairment after mTBI to (1) validate the robustness of QSM and MRF as a routine measurement for quantifying myelin damage and repair process after mTBI on a 3T and clinical 7T animal MRI, (2) characterize white matter axonal and myelin loss and repair process after mTBI by QSM and MWF in the individual level, (3) establish the machine learning based predictive models to predict mTBI patients at a high risk of chronic neurocognitive effects through quantitative MRI measures and patients’ genotyping, and (4) validate the predictive models established in this project by the independent mTBI patient cohort. The verified predictive models may positively impact the diagnostic and therapeutic strategies in mTBI.