Grant: Project Grant

Living with Parkinson's disease (PD) requires patients to manage their disease through self-care. Health technologies can effectively support self-care in PD through tailored health communication using software algorithms that can improve the personalization of information and resources provided. The next generation of self-care technologies will leverage AI techniques to provide interactive educational and decision support mechanisms that empower patients to engage in self-care. The recent advent of large language models (LLMs) offers a promising opportunity to develop a conversational recommender system (CRS) for PD based on natural language processing (NLP) models. However, many technologies fail to fully support self-care because the design process often neglects the complex relationships between technology and individuals in their social, cultural, and economic contexts. More research is needed to explore how CRS can improve the personalization of self-care technologies, particularly through empirical assessments in real-world settings. The overall goal of this study is to co-design an AI-driven self-care technology for people with Parkinson's disease. The specific objectives are (1) to develop a CRS to deliver tailored, accessible, and actionable care tips and resources in French and English; (2) to evaluate the usability, user experience, and likelihood of adoption of the developed AI-based self-care technology. The study will use a co-design approach, in which the perspectives of patients and healthcare providers are considered a central component of the design process, including in-depth collaborative methods as well as evaluation/measurement methods. The research provides a new model for interactive self-care education and decision support, and offers a comprehensive framework for co-designing and implementing AI-driven self-care technologies.