Robotic assisted gait training and gait pattern in children affected with cerebral palsy

Abstract

every foot joint during gait. For this reason a FE model was created and a validated using a dummy model. Patients/materials and methods: Gait analysis in one patient was done with a Vicon MX System and 6 cameras, one AMTI Force plate and one EMED SF pressure measurement platform. The “GaitLowerExtremity” model from the ANYBODY Repository AMMRV1.4.1 and ANYBODY Version 5.1 was used to determine the muscle forces of the lower leg during gait. The muscle forces from the ANYBODY model, the regional reaction forces from the EMED SF platform and the 3D kinematics were used to build up the FE-model. A “standard foot model” was created from a CT scan of one healthy foot. This model was imported in ANSYS Workbench V.14.0 and scaled according to the patient’s standard X-rays. The bony segments were linked by revolute joints. This model was validated by measuring the intra-articular forces at the ankle joint in a dummy model constructed by aluminium and computing the FE Model using data for both analysis from the same patient at 50% stance phase (Figs. 1 and 2). Results: Validation of the FE model shows a good correlation of the measured data in the dummy model and the computed data at the ankle joint in the FE model. Only the moments in z direction (vertical to the floor) showed a difference of 19%. All other differences were in a range between 2 and 8%. Discussion and conclusions: For the dynamic validation of the FEmodel the load during one representative step inmid stancewas applied on the dummymodel. The results of the dummymodel and the FE calculation correlated well. Only the moments in z direction at the ankle joint differed between FE Model and dummy model. The reason is not understood yet and is further analyzed.