Prisma
Overview
Over a quarter of the world’s population is physically inactive, failing to meet global recommendations for weekly physical activity. Mobile health applications present a promising avenue for low-cost, scalable physical activity promotion, but often suffer from small effect sizes and low adherence rates, particularly in comparison to human health coaching. Prisma is a LLM-based conversational agent for personalized health behavior change to encourage physical activity. It extends a purely conversational LLM’s abilities by exposing APIs for querying personal data and for important tasks like goal setting and fact-checking. Unlike existing approaches, this architecture allows the system to integrate various sources of context, including semantic information captured in natural language interaction (e.g., high-level goals and motivation, life circumstances and constraints) and quantitative data about the user’s physiology and behaviors (e.g., biosignals from a wearable). Prisma provides a chat interface and visualization components, as well as a backend server for processing data and managing LLM interactions. We plan to evaluate the conversational agent as a technology probe in a 4-week pilot study with approximately 20 participants.
Features
Data Collection
Prisma imports information from the Apple Health app, improving digital activity trackers for the app. Complementary to this, the Chatbot feature make suggestions to the users based on their questions and commentary, as well as the imported health data. Users can customize which features they allow the app to track.
Survey
Users are recommended to take a daily morning and afternoon survey that will show automatically on the homepage of the app, each of which is a brief multiple choice questionnaire.
Architecture
Faculty and Research Advisors
Student Team
- Dhruv Naik
- Bryant Jimenez
- Caroline Tran
- Evelyn Hur
- Evelyn Song