A fast structured method for single-disease diagnosis information based on medical insurance data

Abstract

The invention discloses a method for rapid structuring of single-disease diagnosis information based on medical insurance data, which structures the diagnosis information in medical big data to construct a single-disease thesaurus; the method includes: extracting the diagnosis information from the medical insurance database; Divide the unstructured text into multiple lexical texts; label the parts of speech of the lexical text; train word vectors; sort them in positive order, and divide them into corresponding word sets; use cosine distance to find the correlation between words; The word list with the most similar standard expressions is used as a standard word list; professionals conduct computer-aided manual verification and repeat it multiple times. The method of the invention can be used to realize the rapid structuring of the individualized diagnosis text data of a single disease, provide technical support for fully and efficiently utilizing the diagnosis information in the medical insurance data, greatly improve the efficiency of data processing and utilization, and speed up the medical big data Conversion promotions.

Description

technical field [0001] The invention provides a fast structuring method for diagnostic data / information about a single disease in a medical insurance database, and belongs to the technical field of medical text processing. Background technique [0002] Medical insurance data (Claims data) is the data generated during the medical insurance business process. The data volume is huge, covering a large-scale group of people, and can completely and truly record the medical treatment information and reimbursement records of the group within a certain period of time. [0003] At present, more and more workers in the medical field are beginning to try to use medical big data for processing and application. Unified and standardized processing, so as to realize the completion of drug evaluation through active monitoring. [0004] However, there are a large number of unstructured texts (mainly diagnostic information) in medical big data. For example, there may be many different express...

AI Technical Summary

Problems solved by technology

[0011] The present invention aims at the non-structural problem of disease diagnosis text or coding (such as multiple myeloma) data existing in the current medical insurance database, and adopts a large and comprehensive matching method different from the previous construction of field databases, and uses single-disease centralized strategy, using the broadest extraction field to include all potential target disease patients into the structured range, and then combined with computer-aided screening and matching, all relevant diagnostic texts of a single disease (such as multiple myeloma) in the medical insurance database For fast and personalized structuring

Images