A modular sensor-based system for the Rehabilitation and Assessment of manipulation

Task-oriented training can reduce motor impairments and improve functional outcomes in subjects affected by stroke. However in most cases this training cannot be performed with sufficient intensity due to limited human and financial resources. Therefore, we are developing a sensor-based System for Independent Task-oriented Assessment and Rehabilitation (SITAR), using an inertial wrist strap, a set of intelligent objects, a marker less camera-based movement tracking system, and dedicated rehabilitative computer games. This paper first presents the Multi-modal Interactive Motor Assessment And Training Environment (MIMATE), a hardware/software platform that serves as a generic basis for designing the intelligent objects and inertial wrist watch. It then describes the design and testing of three intelligent objects - a can, jar and key apparatus - developed to train and assess representative hand functions. Three different tests using these objects were performed on two healthy volunteers, showing how they enable an accurate characterization of the motor behavior during functional tasks.