Magnetic resonance imaging (MRI) can provide several measurements relevant to sport-related concussion (SRC) symptoms; however, most studies to date have used a single MRI modality and whole-brain exploratory analysis approaches, resulting in highly variable findings across studies. A multimodal MRI approach that combines complementary structural and functional information is more likely to yield consistent findings and thus a better understanding of the underlying mechanisms of concussion. The current thesis used multimodal MRI (resting-state functional MRI for functional connection strength and spontaneous activity magnitude, arterial spin labelling (ASL) for cerebral blood flow (CBF), quantitative susceptibility mapping (QSM) for detection of iron accumulation due to inflammation, and diffusion tensor imaging (DTI) for white mater integrity) to investigate SRC in youth aged 12-18 years. The studies consisted of symptom-guided region-of-interest analysis of the cerebellum, basal ganglia, and connecting white matter tracts; a whole-brain exploratory analysis; and a DTI histogram analysis of white matter tracts. Our region-of-interest study demonstrated that youth SRC is associated with acute cerebellar inflammation accompanied by reduced cerebellar activity and cerebellar-parietal connectivity, as well as structural changes of the middle regions of corpus callosum accompanied by functional changes of caudate, all of which resolve with recovery. Whole brain analysis demonstrated that youth SRC is also associated with changes in activity magnitude and CBF in the frontal and occipital poles, in the basal ganglia, and in limbic cortical areas. Residual effects were also present in recovered youth. DTI histogram analysis of white matter tracts revealed an increase in diffusivity skewness within the uncinate fasciculus, a major tract that connects the orbitofrontal cortex to the temporal lobe. The findings revealed by each MRI modality exhibit complementary information on mechanisms of youth SRC, and additional complementary information on concussion mechanisms is gained by using different analysis approaches of multimodal MRI data. Our studies also demonstrates that most changes occur within the first few days post-injury, and hence early MRI is needed to better elucidate brain mechanisms of concussion and symptoms, and to reach the potential of MRI as an objective biomarker for concussion diagnosis, recovery assessment and prediction of outcome.