Traumatic brain injury (TBI) is produced by mechanical loading during road traffic and sporting collisions. TBI is a time-sensitive injury thus requiring rapid post-crash response in road traffic collisions. Moreover, there is growing concern that exposure of the brain to seemingly small loads in sporting can produce long term effects, such as triggering early dementia. Sensing technologies are now used increasingly to measure mechanical loading in both road traffic and sporting. Modern cars are equipped with event data records that measure collision kinematics and use this information to inform post-crash response. Wearables, particularly instrumented mouthguards, are used in sports to measure head kinematics in real-time. However, there is still a gap between the sensor measurements and predicting risk of brain injuries. In this talk, I will present our recent work on utilising sensor data to predict brain injuries. I will explain how we have used the UK road accident in-depth studies (RAIDS) database to prove the feasibility of using the change in the speed of the car (delta-V) to predict TBI severity and pathology. I will also explain how we have used high fidelity computational models of brain biomechanics and machine learning to develop accurate near real-time models for predicting brain response from mouthguard data. I will show how these new advances can help improve brain health in road traffic and sports applications.