Multi-center knowledge graph joint decision support method and system

Abstract

The invention discloses a multi-center knowledge graph joint decision support method and system, which utilizes a medical knowledge graph technology and a block chain technology to realize local semantic reasoning and on-chain result summarizing of clinical data through a mode of combining a local knowledge graph and an on-chain synchronous graph, and therefore, under the condition that original medical data is not out of a hospital, interpretable clinical decision support containing complete clinical evidences of the patient is given based on deductive reasoning and evidence-based medicine by integrating fragmented cross-institution medical data of the patient through the knowledge graph technology. Patient identity information is subjected to anonymous comparison, complete encryption of data in an out-of-hospital link and a matching link is guaranteed, triple node structure information is subjected to coding mapping and asymmetric encryption, the safety of the data in the transmission process is guaranteed, only a participating center with permission can decrypt the data, and the data security and privacy guarantee in the multi-center joint reasoning process are effectively ensured.

Description

technical field [0001] The present invention relates to the technical field of medical knowledge graphs, in particular to a multi-center knowledge graph joint decision support method and system. Background technique [0002] In clinical practice, many patients will see a doctor in multiple hospitals or community hospitals; research shows that 20%-40% of patients will see a doctor in an average of 2 different hospitals within a year. Patients' cross-institution visits will result in fragmentation and isolation of medical data, resulting in incomplete and insufficient patient records in a single hospital; incomplete patient medical data can easily lead to inaccurate clinical decisions by clinicians, resulting in inaccurate clinical decisions. Behaviors such as untimely, inappropriate treatment and repeated medical treatment seriously threaten the quality of medical services and increase the medical burden of the masses. Existing studies have found that incomplete medical data...

AI Technical Summary

Problems solved by technology

[0004] Existing technologies that use multi-center fragmented medical data for decision support have the following problems: (1) Existing technologies only use multi-center medical data in the model training stage, and the trained model can only be used in clinical practice. The patient data of a single hospital is analyzed, and the decision support results still lack the medical data of other hospitals as an auxiliary, and there are still problems in the comprehensiveness and reliability of the generated clinical decisions

(2) The decision support model built based on distributed machine learning, its results are mainly reflected in the form of confidence weight, it cannot give deductive decision support results based on evidence-based medicine, and it is difficult to systematically and comprehensively display decision support related information Patient disease risk factors and clinical evidence, likely to cause low acceptance by doctors

(3) Deductive semantic reasoning technology based on evidence-based medicine. Its distributed algorithm is mainly used for distributed data search and to improve the speed of triplet reasoning; for clinical scenarios of fragmented patient data analysis and application, data security is lacking And privacy protection support, semantic reasoning cannot be performed without summarizing multi-center raw data, and there are still problems in medical data security

Images