EVENTS

Conversational Health Agents: A Personalized LLM-Powered Agent Framework – Workshop on Interactive AI Systems for Digital Therapeutics

Abstract:
Conversational Health Agents (CHAs) are AI-driven systems designed to offer healthcare services, such as assistance, patient self-awareness, and diagnosis. While current CHAs, especially those powered by Large Language Models (LLMs), excel in conversation, they often struggle with analyzing diverse types of data and delivering trustworthy and personalized responses. Our proposed framework, openCHA, addresses these limitations by providing capabilities for CHAs for multi-step problem-solving, multimodal data processing, and access to up-to-date information. To achieve this, openCHA integrates LLMs with external resources, including diverse datasets, knowledge bases, and analytical tools. This framework enables CHAs to process healthcare queries more effectively by analyzing input inquiries, gathering and integrating necessary information, and delivering personalized, context-aware responses. openCHA aims to make CHAs more versatile and responsive to users’ unique healthcare needs.

• Abbasian, M., Azimi, I., Rahmani, A.M. and Jain, R., 2023. Conversational health agents: A personalized llm-powered agent framework. arXiv preprint arXiv:2310.02374
• GitHub repository
• User guide and quick start

Speaker’s bio:
Iman Azimi earned his Ph.D. in Information and Communication Technology from the University of Turku (Finland) in 2019 and completed his M.Sc. in Artificial Intelligence and Robotics at Sapienza University of Rome in 2014. Currently, he serves as the Assistant Director at the UCI Institute for Future Health. His research spans machine learning and deep learning, Generative AI, and biomedical signal processing. He actively contributes to digital health projects focusing on mental health, maternal health, and personalized data analytics. In these projects, Iman develops and applies AI-based techniques to extract personalized insights from physiological, behavioral, and contextual data collected from a variety of sources, including smartwatches. He is passionate about using his expertise and experience to improve people’s quality of life.