Background and Purpose Quantitative Susceptibility Mapping (QSM) using MRI can assess changes in brain tissue structure and composition. This report presents preliminary results demonstrating changes in tissue magnetic susceptibility after sport-related concussion (SRC). Materials and Methods Longitudinal QSM metrics were produced from imaging data acquired on cohort of concussed and control football athletes. 136 QSM datasets were analyzed across three separate visits (24 hours after injury, 8 days post-injury). Longitudinal QSM group analyses were performed on stability-thresholded brain tissue compartments and selected sub-regions. Clinical concussion metrics were also measured longitudinally in both cohorts and compared to the measured QSM. Results Statistically significant increases in white matter susceptibility were identified in the concussed athlete group during the acute (24 hour) and subacute (day 8) period. These effects were most prominent at the 8 day visit, but recovered and showed no significant difference from controls at the 6 month visit. The sub-cortical gray matter showed no statistically significant group differences. Observed susceptibility changes after concussion appeared to outlast self-reported clinical recovery metrics at a group level. At an individual subject level, susceptibility increases within the white matter showed statistically significant correlations with return-to-play durations. Conclusion The results of this preliminary investigation suggest that SRC can induce physiological changes to brain tissue that can be detected using MRI-based magnetic susceptibility estimates. In group analyses, the observed tissue changes appear to persist beyond those detected on clinical outcome assessments, and were associated with return to play duration after SRC.

2 J. Magn. Reson. Imaging 2017;45:51-58.

Purpose: Quantitative Susceptibility Mapping (QSM) reconstruction is a challenging inverse problem driven by poor conditioning of the field to susceptibility transformation. State-of-art QSM reconstruction methods either suffer from image artifacts or long computation times, which limits QSM clinical translation efforts. To overcome these limitations, a deep-learning-based approach is proposed and demonstrated. Methods: An encoder-decoder neural network was trained to infer susceptibility maps on volume parcellated regions. The training data consisted of fabricated susceptibility distributions modeled to mimic the spatial frequency patterns of in-vivo brain susceptibility distributions. Inferred volume parcels were recombined to form composite QSM. This approach is denoted as ASPEN, standing for Approximated Susceptibility through Parcellated Encoder-decoder Networks. ASPEN performance was evaluated relative to several well-established approaches on a gold-standard challenge dataset and on cohort of 200 study subjects. Results: ASPEN provided similar levels of quantitative accuracy compared to the evaluated established approaches on the gold standard ISMRM Challenge dataset, but qualitatively showed marked reductions in streaking artifacts and map blurring. On the large-cohort dataset, ASPEN achieved the highest score compared with other methods in a multi-rater evaluation of streaking artifacts and map resolution. Conclusion:The proposed ASPEN approach can robustly infer susceptibility maps in near real-time on routine computational hardware. This preliminary study establishes ASPEN's parity with existing approaches for quantitative accuracy on a well-curated gold standard dataset and further demonstrates its robustness to streaking artifacts across a large cohort of subjects.