Presentation
Current Applications, Research Gaps, and Future Perspectives in Design of Resilient Patient-Facing Generative AI Applications: A Rapid Review
SessionPoster Session 1
DescriptionGenerative AI (GenAI) technologies, such as ChatGPT, are increasingly being used in patient-facing healthcare applications, raising critical questions about their reliability, safety, and accessibility. This scoping review synthesizes current research on the design and deployment of GenAI applications intended for direct patient use, highlighting emerging use cases, risks, and research gaps. Potential applications include personalized health education, clinical document summarization, and question-answering tools to support patient understanding and engagement. While early studies demonstrate promising outcomes in improving readability and patient comprehension, significant concerns persist regarding misinformation, hallucinations, and lack of safety guardrails. Mitigation strategies, such as retrieval-augmented generation, tailored prompt engineering, and transparency mechanisms (e.g., linking original clinical notes), have been proposed but require further validation. Importantly, few studies have investigated patient preferences, attitudes, or expectations related to these tools. Key open questions include how to preserve privacy, build trust, and ensure equitable access, especially for populations with lower health literacy. This review underscores the urgent need to involve patients in the design and evaluation of GenAI tools to inform future development, regulation, and implementation. As adoption accelerates, prioritizing patient-centered design principles is essential to ensure that GenAI fulfills its promise to expand access to healthcare knowledge safely and effectively.
Event Type
Poster
TimeTuesday, October 14th5:30pm - 6:30pm CDT
LocationRiverside East
Similar Presentations