Knowledge-Infused LLM-Powered Conversational Health Agent: A Case Study for Diabetes Patients

TL;DR

This study tackles the limitation of general-source LLMs in diabetes management by introducing a knowledge-infused conversational health agent (CHA) built on the openCHA framework. The system fuses American Diabetes Association dietary guidelines and Nutritionix nutrition data, plus an analytical module for precise nutrient intake calculations and guideline comparisons, enabling risk-aware dietary guidance. It leverages GPT-3.5-turbo with Tree of Thought planning and consists of Interface, Orchestrator, and External Sources, including a nutrition knowledge base and AI analytics models. In a comparative evaluation against GPT-4 using 100 diabetes-related meal questions, the CHA demonstrates superior accuracy in assessing potential dietary risks, underscoring the value of external knowledge integration and explainable analytics for reliable diabetes management.

Abstract

Effective diabetes management is crucial for maintaining health in diabetic patients. Large Language Models (LLMs) have opened new avenues for diabetes management, facilitating their efficacy. However, current LLM-based approaches are limited by their dependence on general sources and lack of integration with domain-specific knowledge, leading to inaccurate responses. In this paper, we propose a knowledge-infused LLM-powered conversational health agent (CHA) for diabetic patients. We customize and leverage the open-source openCHA framework, enhancing our CHA with external knowledge and analytical capabilities. This integration involves two key components: 1) incorporating the American Diabetes Association dietary guidelines and the Nutritionix information and 2) deploying analytical tools that enable nutritional intake calculation and comparison with the guidelines. We compare the proposed CHA with GPT4. Our evaluation includes 100 diabetes-related questions on daily meal choices and assessing the potential risks associated with the suggested diet. Our findings show that the proposed agent demonstrates superior performance in generating responses to manage essential nutrients.

Figures (3)

• Figure 1: An overview of openCHA abbasian2023conversational framework.
• Figure 2: LLM-based CHA for diabetes management enabled by the openCHA framework.
• Figure 3: A sample question and responses from the proposed CHA and GPT4. The green text indicates matching the identified risk with the ground truth, the red text indicates mismatching.