Towards Personalized Interaction and Corrective Feedback of a Socially Assistive Robot for Post-Stroke Rehabilitation Therapy

Abstract

The research of a socially assistive robot has a potential to augment and assist physical therapy sessions for patients with neurological and musculoskeletal problems (e.g. stroke). During a physical therapy session, generating personalized feedback is critical to improve patient's engagement. However, prior work on socially assistive robotics for physical therapy has mainly utilized pre-defined corrective feedback even if patients have various physical and functional abilities. This paper presents an interactive approach of a socially assistive robot that can dynamically select kinematic features of assessment on individual patient's exercises to predict the quality of motion and provide patient-specific corrective feedback for personalized interaction of a robot exercise coach.

Enabling AI and Robotic Coaches for Physical Rehabilitation Therapy: Iterative Design and Evaluation with Therapists and Post-stroke Survivors

“…Motion tracking techniques can be further developed into various applications for better rehabilitation experiences for a patient. These applications include virtual reality [39], intelligent coaching systems [22], and assistive robots [11,25,26].…”

“…Exoskeleton robots have been explored to augment patient's weak body limbs and induce a passive motion for rehabilitation [26]. AI [22] or robotic coaches [11,25,47] can guide patient's rehabilitation through automatically monitoring patient's exercises [23] and providing feedback on whether a patient performs well-being-related or rehabilitation exercises correctly or not [11,17,25,29,41]. As prior work has demonstrated the benefit of physical embodiment to improve the engagement in physical exercises [11], we decided to further explore research on socially assistive robotics.…”

“…In addition, these systems can assist patient's engagement in well-being-related or rehabilitation exercises through social interaction (e.g. providing encouragement [11,25,29]). In this paper, we focus on a system that leverages AI techniques and robotic embodiment to coach exercises and refer it as an AI and robotic coach.…”

Enabling AI and Robotic Coaches for Physical Rehabilitation Therapy: Iterative Design and Evaluation with Therapists and Post-Stroke Survivors

“…A large body of work on socially assistive robotics has focused on a specific technical improvement (e.g. improving a technique of automated assessment [48,27]) or the effect of a particular design variable (e.g. physical embodiment [13], matching the style of interaction with user's personality [49], the usage of comparative feedback [47]).…”

“…Instead, we implemented a system to check the occurrence of a compensated motion (e.g. leaning trunk to the side) [27]. Once a compensated motion is detected multiple times, we utilized this detection to trigger an interactive dialogue and confirm the user's subjective status.…”