Medical big data multi-center integration platform and method

Abstract

The invention discloses a medical big data multi-center integration platform and method. Data of each data branch center server are accessed to a data center server and subjected to quality assessment, and a next step is entered if the quality assessment passes; if the quality assessment fails, the data center server feeds back the conclusion of failure to each corresponding data branch center server; the quality assessment includes the assessment of data integrity, data repetition rate, data bias and data size; the data center server establishes and maintains standard variables and a standarddictionary, meanwhile, preprocesses the data according to the standard variables and the standard dictionary; the data standardization processing includes standardization of the variables and standardization of data values; a one-to-one mapping relationship between data variables of the data branch center servers and the standard variables of the data center server is established by means of a similarity matching algorithm and manual auditing; and the data standardized by the data center server are utilized.

Description

technical field [0001] The invention relates to a medical big data multi-center integration platform and method. Background technique [0002] There are following problems in the prior art to be solved: [0003] First, the amount of data is huge and complex; the data includes physical examination data from dozens of medical examination centers, basic public health services in multiple regions, government data on women of childbearing age, clinical data from multiple tertiary hospitals, and multiple specialized medical records. Disease data, such as: psychiatric data, glioma and other major disease databases, each data source stores a large amount of data, and the data format of each data source varies greatly; [0004] Second, the disadvantages of traditional data collation. Traditional data collation is aimed at a single database, which consumes a lot of manpower and material resources to organize data, statistical analysis, and discover valuable scientific research result...

AI Technical Summary

Problems solved by technology

[0003] First, the amount of data is huge and complex; the data includes physical examination data from dozens of medical examination centers, basic public health services in multiple regions, government data on women of childbearing age, clinical data from multiple tertiary hospitals, and multiple specialized medical records. Disease data, such as: psychiatric data, glioma and other major disease databases, each data source stores a large amount of data, and the data format of each data source varies greatly;

[0004] Second, the disadvantages of traditional data sorting, traditional data sorting is aimed at a single database, which consumes a lot of manpower and material resources to sort out data, statistical analysis, and discover valuable scientific research results

[0008] Sixth, the relationship between scientific research and collation: as we all know, data collation is the premise of scientific research statistics, but there is an embarrassing problem. It is very likely that the research indicators required by scientific research cannot be satisfied in the collated database. For example, our scientific research needs to study The indicator "non-alcoholic fatty liver", in the process of general data collation, the physical examination indicators include drinking alcohol and whether the ultrasound diagnosis is fatty liver. The type of fatty liver needs to be defined by the researchers themselves, and the original data needs to be sorted out again

Images