Non-accidental head injury in infants has far-reaching medical and legal implications. The symptoms typically involve a `triad' including subdural haemorrhage (bleeding in the region between the skull and the brain), retinal haemorrhage (bleeding of the retina at the back of the eye) and brain swelling. However legal investigation is difficult as the cause and effect of the injury are not always easy to distinguish. This project will use mathematical modelling to elucidate the origin(s) of one of these symptoms, retinal haemorrhage, constructing a novel new mechanical framework capable of predicting the onset and severity of retinal bleeding following a traumatic brain injury, in an attempt to correlate the pattern of bleeding to the injury sustained and thus aid clinical decision making. Such a traumatic brain injury will create a steep pressure wave (like a tsunami) which will propagate through the blood vessels in the retina, causing some of them to burst and hence bleeding (haemorrhage). This project will examine the mathematical properties of such a pressure wave as it spreads across blood vessel junctions and hence across an entire networks of arteries and veins. This project will be to examine and quantify how the final pattern of blood vessel bursting (ie bleeding) correlates to the type of injury sustained. Such a correlation will be of enormous value to doctors and legal practitioners making difficult decisions about the origin of injuries in infants.