A general knowledge graph dynamic construction and reasoning method and system based on multi-source medical data fusion

By constructing and reasoning a general practice knowledge graph through the fusion and dynamic updating of multi-source medical data, the problem of insufficient interpretability in the fusion, dynamic updating, and reasoning of multi-source heterogeneous data in general practice knowledge graphs is solved, thus realizing efficient and interpretable auxiliary diagnosis and treatment support of general practice knowledge.

CN122436264APending Publication Date: 2026-07-21SHANGHAI YICHI SHIJIE DIGITAL TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610513814.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-04-17
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

Existing medical knowledge graph construction technologies are ill-suited to the complex needs of general practice medicine, which involves cross-specialty and multi-disease associations. They suffer from weak multi-source heterogeneous data fusion capabilities, lack of dynamic updates, insufficient interpretability of reasoning, and difficulty in adapting to the hierarchical diagnosis and treatment logic of general practitioners.

Method used

By employing a multi-source medical data fusion approach, a comprehensive knowledge graph is constructed through data preprocessing, knowledge fusion, dynamic updating, and reasoning steps. This includes entity alignment, conflict resolution, knowledge completion, causal subgraph construction, and a hybrid reasoning engine. This achieves efficient fusion and dynamic updating of multi-source data and generates accurate reasoning results through the combination of symbolic reasoning and deep learning.

Benefits of technology

It has achieved a knowledge graph that covers all disciplines and has strong timeliness and interpretability, which significantly reduces the workload of manual construction and updating, adapts to the diagnosis and treatment process of general practitioners, and provides continuously optimized auxiliary decision-making capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122436264A_ABST
    Figure CN122436264A_ABST
Patent Text Reader

Abstract

The application discloses a kind of general knowledge graph dynamic construction and reasoning method and system based on multi-source medical data fusion, comprising: obtaining multi-source medical data and respectively carrying out the entity relationship extraction of structured, unstructured text and visual data;Based on general medical knowledge framework, entity alignment, conflict resolution and knowledge completion are carried out, and general knowledge graph is constructed;Listen to new medical data, and the graph is iteratively updated by incremental extraction and double verification mechanism;Causal subgraph containing strong causal relationship edge is constructed, and reasoning result is generated by the hybrid reasoning engine of the fusion of symbolic reasoning and deep learning;Based on doctor feedback, adjust reasoning rule weight and causal relationship determination.The application solves the problems of multi-source data fusion difficulty, dynamic update deficiency and insufficient reasoning interpretability in the prior art, realizes the comprehensive coverage, time-effective update and precise and interpretable clinical auxiliary reasoning of general knowledge graph.
Need to check novelty before this filing date? Find Prior Art