ICU electronic medical record knowledge graph construction method and system and storage medium

Abstract

The invention discloses an ICU electronic medical record knowledge graph construction method and system and a storage medium. The method comprises the steps: creating an ICU patient feature data set;carrying out the knowledge extraction of facts and reasoning through the preprocessed ICU patient feature data set and an association rule mining algorithm, and expressing knowledge in a triple form;constructing a knowledge graph based on facts according to the existing facts reflected by the feature data set, and mapping inference knowledge between many-to-many entities into a graph structure onthe basis of the knowledge graph based on facts to complete ICU electronic medical record knowledge graph construction based on facts and inference; generating a low-dimensional dense vector mapped by each patient feature by utilizing a graph embedding method based on facts and reasoning; calculating a similarity matrix between the low-dimensional dense vectors of each patient, and returning Topk similar electronic medical records of the given patient according to a similarity sequence. According to the method, the ICU electronic medical record knowledge graph is constructed on the basis offacts and reasoning, so that the completeness of the ICU electronic medical record knowledge graph is improved.

Description

technical field [0001] The present invention relates to the technical field of knowledge map of medical electronic medical records, in particular, to a method, system and storage medium for constructing a knowledge map of ICU electronic medical records. Background technique [0002] In terms of knowledge map construction, since ICU electronic medical records involve a large amount of professional knowledge in the medical field, which is different from knowledge representation in other fields, experts are often required to add prior knowledge according to the rules, but manual knowledge extraction is heavy and time-consuming. , it is difficult to implement manually for large databases. There are also studies to improve the knowledge graph by introducing other knowledge bases in the medical field, but there are diversity and heterogeneity among different knowledge bases, and the newly introduced entities and relationships usually do not match well with the entities and relatio...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the drawbacks in the above-mentioned prior art, the present invention provides a method and system for constructing a knowledge map of ICU electronic medical records based on facts and reasoning. The knowledge map of ICU electronic medical records is constructed based on facts and reasoning, which can cover existing The explicit information of known facts can also be used to expand the map through the implicit information mined by the association rule algorithm, so as to increase the completeness of the knowledge map of ICU electronic medical records, and at the same time avoid the problems caused by manual introduction of rules or fusion of multiple knowledge bases

Images