Infectious disease monitoring and early warning method and system based on large language model and knowledge graph
By combining large language models with knowledge graphs, a port spatiotemporal knowledge graph is constructed and a metapopulation graph transformation neural network is used to solve the problems of heterogeneity and dynamic modeling of multi-source data, realize efficient infectious disease monitoring and early warning, and provide accurate risk assessment and intervention strategies.
Patent Information
- Application Number
- CN202610864371.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-16
- Publication Date
- 2026-07-17
AI Technical Summary
Existing infectious disease monitoring and early warning systems struggle to effectively integrate multi-source heterogeneous data in port scenarios, lack the ability to dynamically model cross-regional, cross-species, and cross-modal transmission risks, and are prone to misjudging or missing transmission signals. They also lack the ability to adaptively characterize the dynamic relationships between population movement, animal habitats, and logistics routes.
By combining a large language model with a knowledge graph, cross-modal feature alignment is achieved through a dynamic time lag masking mechanism, and a spatiotemporal knowledge graph covering people, animals, goods, and the environment is constructed. A meta-population graph transformation neural network is used for forward graph convolutional propagation, and a multi-agent cluster is activated to perform counterfactual inference to generate structured disposal instructions.
It has achieved efficient fusion and feature alignment of multi-source data, improved the ability to quickly trace the transmission path of infectious diseases and assess risks, enhanced the timeliness and accuracy of early warning, and provided support for scientific intervention strategies.
Smart Images
Figure CN122417461A_ABST