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

• DOI: 10.1007/s12369-022-00883-0
• Publication Date: 2022-07-23T18:13:32Z
• Authors: Min Hun Lee; Daniel P. Siewiorek; Asim Smailagic; Alexandre Bernardino; Sergi Bermúdez i ...
• Keywords: Stroke

Artificial intelligence (AI) and robotic coaches promise the improved engagement of patients on rehabilitation exercises through social interaction. While previous work explored potential automatically monitoring for AI coaches, deployment these systems remains a challenge. Previous described lack involving stakeholders to design such functionalities as one major causes. In this paper, we present our efforts eliciting detailed specifications how could interact with guide patient's in an effective acceptable way four therapists five post-stroke survivors. Through iterative questionnaires interviews, found that both survivors appreciated benefits achieve more systematic management improve their self-efficacy motivation therapy. addition, evaluation sheds light several practical concerns (e.g. possible difficulty interaction people cognitive impairment, system failures, etc.). We discuss value early involvement interactive techniques complement but also support personalized therapy session better exercise coaches.