Multimodal Interfaces to Improve Therapeutic Outcomes in Robot-Assisted Rehabilitation

The paper presents the developing of a new robotic system for the administration of a highly sophisticated therapy to stroke patients. This therapy is able to maximize patient motivation and involvement in the therapy and continuously assess the progress of the recovery from the functional viewpoint. Current robotic rehabilitation systems do not include patient information on the control loop. The main novelty of the presented approach is to close patient in the loop and use multisensory data (such as pulse, skin conductance, skin temperature, position, velocity, etc.) to adaptively and dynamically change complexity of the therapy and real-time displays of a virtual reality system in accordance with specific patient requirements. First, an analysis of subject's physiological responses to different tasks is presented with the objective to select the best candidate of physiological signals to estimate the patient physiological state during the execution of a virtual rehabilitation task. Then, the design of a prototype of multimodal robotic platform is defined and developed to validate the scientific value of the proposed approach.