Clinical data automatic generation method and system based on causal relationship mining

Abstract

The invention discloses an automatic clinical data generation method and system based on causal relationship mining, and the method comprises the steps: carrying out the classification statistics of the information of a patient in a hospital from a collection source of medical data according to the hospital admission and discharge process of the patient, neglecting the redundant information generated in the hospital admission process of the patient, and carrying out the statistical analysis of the information. The hospital admission process of a patient is sorted into four items of illness state, examination, illness and medicine, the four items are integrated into a unified data set, then data in the data set is screened for the first time according to the personalized requirements of a user, then the dependency relationship and dependency direction of all columns of data are determined through an algorithm, a directed acyclic graph of the complete part of the selected data is drawn, and the directed acyclic graph of the complete part of the selected data is obtained. Therefore, causal association among the data is found, and then the part in which the user is interested is selected from the causal graph for virtual generation of the causal association, so that the problem that the generated virtual data are not closely associated is solved.

Description

technical field [0001] The invention relates to the technical field of medical information, in particular to a method and system for automatic generation of clinical data based on causal relationship mining. Background technique [0002] Through the integration, analysis and mining of clinical data of individual patients, a good health model can be established for individual patients, and a precise and accurate disease prevention plan and treatment plan can be provided for patients. Once the clinical data of individual patients is integrated and collected and researched by the institution, it can assist in the development of medical software, the development of new drugs, etc., which will greatly assist the medical industry, but these clinical data often contain sensitive individuals. Information, once the private information is leaked, it may have a certain negative impact on the lives of patients, so the medical department will have concerns when using this data. [0003]...

AI Technical Summary

Problems solved by technology

However, due to the black-box nature of the neural network, it is impossible to properly explain the generated virtual data, and the medical field places more emphasis on the interpretability of the source of results than other industrial and mechanical fields, which leads to low applicability of the model , not of general use value

Images