Medical examination knowledge graph construction method based on LIS

Abstract

The invention discloses a medical examination knowledge graph construction method based on LIS. The method comprises the steps of concept layer design, instance layer learning and application of an LIS knowledge graph. Specifically, a real record in an LIS system is used as a knowledge base, a concept set in the knowledge field is designed, and the form expression of concepts and concept relationsis emphasized. Medical examination facts matching the conceptual layer are extracted from a large number of LIS system records. Medical examination entities and relations are extracted and the medical examination entities and relations are sorted in a triple form, thereby obtaining a final medical examination knowledge map. The medical examination knowledge graph is applied to two aspects of examination query and reasonable examination. A medical examination knowledge graph is constructed based on real data in medical examination documents and LIS, so that a general medical knowledge graph isperfected, and the medical examination knowledge graph can better serve high-level artificial intelligence medical application. Meanwhile, as an auxiliary query tool of the doctor, the work efficiency of the doctor is improved, and the application scene of the LIS system is developed.

Description

technical field [0001] The present invention relates to the technical field of knowledge map creation, in particular to a method for constructing a medical test knowledge map based on LIS. Background technique [0002] The laboratory information system (LIS) realizes the information collection, storage, processing, transmission and query of clinical laboratories, provides analysis and diagnosis support, assists laboratory doctors in preprocessing test application forms and specimens, and automatically collects test data Or directly input, test data processing, etc. LIS contains the basic information of patients, various test data and test diagnosis results, etc., and records medical entities and their associations, such as patient entities, basic information entities, test entities, diagnostic entities and other entities and their relationships, which include With a large number of medical test facts. Therefore, in-depth mining of medical inspection factual knowledge in th...

AI Technical Summary

Problems solved by technology

However, most of the existing medical knowledge graphs obtain knowledge from public medical literature and medical dictionary manuals, and do not involve the construction and application of medical inspection knowledge graphs based on real inspection data of the LIS system.

Images