HCC: Advancing Human-Centered AI in Personal Health Informatics Tools: Support for Planning, Tracking, and Reflection in Management, Information, and Tracking Goals

Personal health informatics offers a transformative approach to supporting health through systematic data collection and personalized information. Despite this promise, current personal health informatics tools often fall short because they do not support the wide range of goals people bring to self-tracking. This project therefore advances human-centered artificial intelligence (AI) for personal health informatics, aiming to help people better understand and manage their goals and helping to develop tasks that meet those goals. It develops methods and tools that leverage human-centered AI to support goal-directed planning, tracking, and reflection. It also investigates how AI can support a broader range of health goals while enabling deeper personalization within those goals. The project will research new methods and tools advancing human-centered AI in personal health informatics tools, leveraging representation of goal context alongside a tracking plan. Development will be based in several tools, including a mobile app, a web / tablet tool, and backend services. This will leverage strengths of different forms (e.g., mobile data capture, larger devices for analysis and collaboration) and integrate with mobile health data frameworks. The research focus is on effective human-centered AI in the context of personal health informatics tools. An initial focus will be human-centered AI support as individuals define goals and translate them into tracking plans. This will generate new understanding of how human-centered AI can enable goal-directed support in goal elicitation, visualizations and analyses, and interactive alignment. The project will also advance AI support for goal evolution, including through mixed-initiative techniques, the use of historical goal context, and the management of goal lapsing and resumption. The project will further investigate AI support for goal-related collaboration, including patient collaboration with providers and with support networks. The research will be conducted through field deployments, developing community-engaged recruitment through a personal health informatics helpdesk. Mixed-method evaluations will thus examine how human-centered AI supports participants as they navigate their own health concerns and questions. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.