NCT07308366

Research on an Intelligent Health Recommendation System for Chronic Disease Comorbidity Integrating TCM

Research on an Intelligent Health Preservation Recommendation System for Chronic Disease Comorbidity Based on the Integration of Traditional Chinese Medicine Constitution Database and Multimodal Large Language Models

• Status: Not Yet Recruiting
• Phase: —
• Study type: Observational
• Enrollment: 195 (estimated)
• Sponsor: The Fourth Affiliated Hospital of Zhejiang University School of Medicine · Academic / Other
• Sex: All
• Age: —
• Healthy volunteers: Not accepted

Summary

1. Construct a Traditional Chinese Medicine (TCM) constitution database, clarify the distribution patterns of TCM constitution in populations with comorbid "three-high" conditions (hypertension, hyperlipidemia, and hyperglycemia) and their associations with metabolic indicators. Establish a "constitution-comorbidity-metabolism" relationship model to provide a basis for personalized intervention and the development of an AI platform. 2. Develop the AI-HEALS system by integrating the TCM constitution database with multimodal large language models. This system will generate personalized intervention plans and provide intelligent interactive Q\&A capabilities to enhance patient intervention adherence. 3. Evaluate the clinical application effectiveness of the AI-HEALS system, explore the relationship between changes in constitution and intervention outcomes, and validate the TCM intervention pathway of "regulating constitution to promote health." This will provide both theoretical and practical guidance for the dynamic regulation and precise intervention of TCM constitution.

Detailed description

This project combines Traditional Chinese Medicine (TCM) constitution theory with large language models (LLMs) through interdisciplinary integration, constructing a dynamically empowered intelligent health recommendation system for TCM. It promotes the deep integration of the "treatment based on constitution differentiation" concept with artificial intelligence. The significance of this research is mainly reflected in the following two aspects: At the theoretical level, this study helps expand the knowledge representation and computational modeling methods of TCM constitution theory within the framework of modern artificial intelligence. It advances the application and transformation of the TCM concept of "preventive treatment" in big data and intelligent reasoning scenarios, provides new perspectives for research on the mechanisms linking TCM constitution and chronic disease comorbidities, and fosters cross-integration between TCM theoretical systems and modern medical information science. At the practical level, the research relies on real clinical data and multimodal AI models to establish a structured, standardized TCM constitution database. It develops a health education system with individualized identification, intelligent recommendation, and dynamic intervention functions, suitable for personalized management and early warning in populations with chronic disease comorbidities. The project outcomes will help enhance individual health literacy and quality of life, alleviate the burden of chronic diseases, promote the practical application of TCM in primary healthcare services and digital medicine, and demonstrate significant social value and broad prospects for widespread adoption.

Conditions

• TCM Constitution Theory

Interventions

Type	Name	Description
OTHER	Multimodal AI Models	Construct a Traditional Chinese Medicine (TCM) constitution database, clarify the distribution patterns of TCM constitution in populations with comorbid "three-high" conditions (hypertension, hyperlipidemia, and hyperglycemia) and their associations with metabolic indicators. Establish a "constitution-comorbidity-metabolism" relationship model to provide a basis for personalized intervention and the development of an AI platform.

Timeline

• Start date: 2025-12-20
• Primary completion: 2028-12-30
• Completion: 2028-12-30
• First posted: 2025-12-29
• Last updated: 2025-12-29