Diabetes Prevention Chatbots to Help Dietary Self-management for the Chinese Population

Area of Science:

• Artificial Intelligence in Healthcare
• Public Health
• Digital Health Interventions

Background:

• Diabetes prevalence is a growing concern in China.
• Existing diabetes prevention methods have limitations in personalization and cultural relevance.
• There is a need for innovative strategies to enhance diabetes management in the Chinese population.

Purpose of the Study:

• To propose an AI-driven web application for personalized diabetes prevention tailored to the Chinese population.
• To improve the effectiveness and accessibility of diabetes prevention strategies in China.
• To demonstrate the feasibility of AI in enhancing early disease prevention.

Main Methods:

• Fine-tuning a large language model (LLM) using a custom dataset.
• Dataset construction based on a self-defined diabetes prevention guideline focusing on Chinese lifestyle and dietary habits.
• Development of an interactive web application enabling real-time user consultation and data download.

Main Results:

• The AI application offers personalized suggestions for diabetes prevention.
• The system addresses the lack of personalization and cultural relevance in existing tools.
• Demonstrated feasibility of AI for enhancing early prevention strategies in healthcare.

Conclusions:

• AI-driven applications can significantly enhance diabetes prevention strategies.
• The developed application bridges gaps in current prevention tools by offering culturally relevant, personalized advice.
• Future implications include expanded features, community integration, and healthcare system linkage for improved Chinese diabetes prevention.