Talk2Care: Facilitating Asynchronous Patient-Provider Communication with Large-Language-Model

• DOI: 10.1609/aaaiss.v4i1.31785
• Publication Date: 2024-11-08T11:45:08Z
• Authors: Xuhai Xu; Bingsheng Yao; Ziqi Yang; Shao Zhang; Ethan Rogers; Stephen Intille; Nawar Shara; Guodong Gao; Dakuo Wang
• Keywords: H.5.2; FOS: Computer and information sciences; Computer Science - Computation and Language; Artificial Intelligence (cs.AI); Computer Science - Artificial Intelligence; I.2.7; H.5.2; I.2.7; Computer Science - Human-Computer Interaction; Computation and Language (cs.CL); 68U35; 3. Good health; Human-Computer Interaction (cs.HC)

Despite the plethora of telehealth applications to assist home-based older adults and healthcare providers, basic messaging phone calls are still most common communication methods, which suffer from limited availability, information loss, process inefficiencies. One promising solution facilitate patient-provider is leverage large language models (LLMs) with their powerful natural conversation summarization capability. However, there a understanding LLMs' role during communication. We first conducted two interview studies both (N=10) providers (N=9) understand needs opportunities for LLMs in asynchronous Based on insights, we built an LLM-powered system, Talk2Care, designed interactive components groups: (1) For adults, leveraged convenience accessibility voice assistants (VAs) conversational interface effective collection. (2) health LLM-based dashboard summarize present important based adults' conversations VA. further user evaluate usability system. The results showed that Talk2Care could process, enrich collected considerably save providers' efforts time. envision our work as initial exploration capability intersection interpersonal