Clinical test item data processing method and device and electronic device

Abstract

The invention provides a clinical test item data processing method and device and an electronic device, relates to the medical technical field, is to solve the problem of low accuracy of risk prediction in the prior art and provides a comprehensive set of techniques and concepts to discover the relationship among diseases, clinical test and genes. The data processing method comprises the followingsteps: obtaining original data of clinical test items; based on a prediction model based on support vector regression, supplementing missing data in the original data, and obtaining transform data ofthe clinical test items; and based on the transform data, generating correlation data of the clinical test items according to correlation between the different clinical test items. The method can solve three problems, that is, missing data, unbalanced distribution and uncertain data annotation in EHR data, realizes disease risk prediction through an association network between the clinical test items, and improves disease risk prediction accuracy.

Description

technical field [0001] The invention relates to the field of biomedical technology, in particular to a data processing method, device and electronic equipment for clinical testing items. Background technique [0002] With the rapid development of biomedical informatics and computer systems, electronic health records (EHR) are widely used. In addition, with the national health care plan, more and more people participate in regular physical examinations and other medical activities, making EHR a big data resource for biomedical research and health care services. [0003] Custom statistical data mining and machine learning techniques have been applied to EHR and to address difficulties encountered in biomedical and healthcare big data. However, in the field of EHR big data, there are three problems that need to be paid attention to. First, EHR data contains missing values ​​caused by human errors; second, EHR data lacks in-depth and unified annotation, that is, for the same A...

AI Technical Summary

Problems solved by technology

However, in the field of EHR big data, there are three problems that need to be paid attention to. First, EHR data contains missing values ​​caused by human errors; second, EHR data lacks in-depth and unified annotation, that is, for the same As a result, different patients may get different annotations; third, EHR data has a small number of naturally occurring abnormal events (uneven distribution of data) that lead to biased analysis results

[0004] At present, although many advanced data analysis technologies and systems have been developed to manage electronic medical record data, each technology or system can only be applied to one aspect or a few simple aspects, such as OpenMRS and dhis2 technologies can only be applied in Databases, data integration, and simple statistical reporting aspects

However, in the application of EHR data, for a large number of high-risk diseases, there is no technology or system to solve the above-mentioned three problems that need to be paid attention to in the large-capacity electronic medical record data

[0005] To sum up, the existing data analysis technology and system in the application of EHR data have the problem of low accuracy of disease risk prediction

Images