Method and device for judging attributes of medical entities, and related product

Abstract

The invention discloses a method and device for judging attributes of a medical entity and a related product. The method comprises the steps of firstly, applying a label generation model to process amedical record text to obtain a label string; and then, according to the attribute tags in the tag string and the sorting of the attribute tags, obtaining the attribute tags of the medical entities inthe medical record text. The attribute label of the medical entity can be used for judging the attribute of the medical entity. The first model in the label generation model dynamically generates thevector of the character according to the actually input medical record text, and the label generation model combines the context information of the character when generating the attribute label of the character, so that the accuracy of the attribute label of the character is improved. The second model is obtained by training through a deep learning method, the label string can be generated according to the vector string output by the first model, and it is guaranteed that label attributes in the label string orderly correspond to characters in the medical record text. Therefore, compared withthe prior art, the accuracy of medical entity attribute judgment is improved.

Description

technical field [0001] The present application relates to the technical field of data processing, in particular to a method, device and related products for discriminating attributes of medical entities. Background technique [0002] Entities refer to things that exist objectively and can be distinguished from each other. Medical entities refer to things that have meaning and are different from each other in the medical field. For example, cough, fever, heart disease, etc., all belong to medical entities. Medical entities widely exist in medical record texts. With the development of informatization in the medical field, the value of medical big data has become increasingly prominent, and the data standardization of electronic medical records has become the key to the exchange of medical data. In order to realize the data standardization of electronic medical records, it is necessary to carry out structured extraction of medical record texts first. When performing structu...

AI Technical Summary

Problems solved by technology

Among them, the regularization-based method uses a dictionary to directly judge, and the error rate is high; while the sentence classification method is used, when the sentence is too long, the context information is easily lost in the attribute discrimination of the medical entity, so the error rate is also high

Images