MOCAP and digital human model techniques to identify gait deviations in patients with lower limb prosthesis

The paper refers to the design of lower limb prostheses adopting a fully virtual approach. In particular, it addresses issues arising when the prosthesis needs to be virtually tested, i.e., the gait analysis of the patient’s avatar wearing the prosthesis. The underlying idea is to use a Digital Human Modeling system to virtually simulate and test the patient’s movement and postures. Extrapolating geometric parameters for each body segment or joint in any time frame of a step permits to identify gait deviations. To this end, we have defined a map connecting each deviation in the amputees’ gait with the only parameters sensible to the said deviation and with a range of referring values. Therefore, it is possible to create a direct flow of information going from the patient gait to the automatic identification of one or more gait deviation, which will be the base for the identification of causes and remedies. To validate our approach, we set up a low cost portable Motion Capture (Mocap) system to acquire the subjects’ gait (both able and disable people) and, by means of an ad hoc data exchange module, gait data have been passed to a detailed biomechanical DHM tool where the patients’ avatar is simulated.