Method and device for label processing of clinical data

Abstract

The present application discloses a clinical data label processing method and device. The method determines a first data item in a clinical data record table in response to a first label generation instruction, where the first data item includes a second data item corresponding to a target operation and potential target data items, determine the target data corresponding to the first label based on the data generation time of the second data item and the potential target data item, generate a label database based on the basic information of the first label and the target data corresponding to the first label, The basic information of a tag includes the tag name and tag weight. The tag weight is used to represent the confidence of the tag. The tag weight is dynamically updated based on the update frequency of the target data corresponding to the tag. In response to the data query instruction, the target tag is determined and based on The current label weight of the target label displays the query results, which can improve the efficiency of clinical data labeling and the accuracy of labels, and ensure the accuracy of clinical data query results.

Description

technical field [0001] The present application relates to the field of data processing, and in particular, to a method and device for labeling clinical data. Background technique [0002] In the era of big data, data has changed the service model of the traditional medical industry while showing changes in mass, diversification and value. How to obtain and filter valuable information from massive clinical data at PB level or even EB level is a major challenge for the hospital information department. By constructing patient labels and supporting precise scientific research services, it is an effective solution to the above challenges. [0003] However, unlike Internet labels, the clinical data generated by hospital informatization includes unstructured data, such as postoperative bleeding, preoperative yellowing methods, and secondary operations. For the labeling of these unstructured data with time attributes, the existing technical solution is to manually confirm the time...

AI Technical Summary

Problems solved by technology

For the labeling of these unstructured data with time attributes, the existing technical solutions are based on hospital clinical data, manually confirming the time when the data is generated, and then distinguishing between preoperative and postoperative data and labeling (that is, labeling). Not only is it time-consuming and labor-intensive, but the accuracy of the generated labels cannot be guaranteed. Correspondingly, the accuracy of the query results of subsequent clinical data cannot be guaranteed.

Images