The core objective of this project is to improve care of children with traumatic brain injury and concern for abuse by using machine learning to recognize new patterns of brain injury that are associated with abuse and with poor outcomes. The team has developed cutting-edge MRI-based imaging methods, and are well-suited to identify new patterns of brain injury that can be used to improve diagnosis and predictions for clinical care. The core methods of this project are to combine clinical and imaging data for approximately 500 children with severe brain injury, and to use traditional and machine-learning analysis to identify characteristics associated with abuse and with poor outcomes. In the first year, they have accomplished the clinical objectives ahead of schedule. Having planned to collect data on approximately 250 eligible subjects, they now anticipate complete data collection on nearly 500 subjects. They obtained Institutional Review Board approval ahead of schedule and have added all new investigators to the research team. Though they were unexpectedly faced with the loss of our statistical and machine-learning partners, they were able to establish improved partnerships within the University, including a new partnership with Terri Lewis, another MINDSOURCE investigator. They have identified imaging studies for the nearly 500 participants, and have established a confidential imaging export protocol, and processes to compare studies across age and size. They anticipate beginning core machine-learning analyses early in the second year of the project. The work has been largely unaffected by the COVID pandemic, and they are able to continue all research activities despite a shutdown of the Anschutz Medical Campus.