Shaken Baby Syndrome is a collection of injuries that have been associated with the violent shaking of an infant or small child. These injuries can then lead to serious brain damage or even death. It is therefore important to identify the exact mechanism that leads from the shaking to the observed injuries, but little experimental work has been done in this area.
The first part of this study was designed to identify if a correlation exists between the physical characteristics of a person shaking a crash test dummy (CRABI) and the resulting accelerations and jerks associated with the motion of the dummy’s head. This was done by placing a three axis accelerometer in the head and two in the body (one in the chest and one in the groin) of a median twelve month old male dummy to determine the acceleration of the head and body. In particular, the relative angular acceleration and jerk of the head relative to the body was determined, since it was felt to be a better predictor of brain damage than would be the absolute linear acceleration of the head.
Similar work has been done in the past; however that study only considered the absolute acceleration of the head, and in only one direction. Since the present study allows the attitude of the head to be determined, a true relative angular acceleration of the head relative to the body was found. Consequently, it was found that no strong correlation existed between the absolute linear acceleration and any body characteristic, however a correlation (R2) of 0.6 was found to exist between the body weight of the shaker and the maximum angular jerk of the dummy’s head relative to its body, as compared to only a correlation of 0.5 when the shaker’s body weight was compared to the absolute linear acceleration of the head.
A two dimensional dynamic simulation was also developed that modelled the behavior of a child crash test dummy. The model included the legs, torso, and head of the dummy, and the elastic behavior of the neck. The model was created to allow the associated accelerations and jerks to be determined for inputs of various magnitude and temporal profiles. The model was then validated by comparing the simulation results to the test results obtained from the experimental study described above.