International Classification of Diseases code recommendation method, system, corresponding equipment and storage medium

Abstract

This application discloses an International Classification of Diseases coding recommendation method, system, corresponding equipment and storage medium, wherein the method includes: mapping the ICD name text to be recommended for ICD coding to the corresponding entity in a pre-formed entity dictionary; mapping the mapped entity Convert to a multidimensional word vector; input the multidimensional word vector and ICD name text to the trained deep learning model; use the top m ICD suborder codes with the highest probability in the output of the deep learning model as recommended ICD suborder codes; where, The entity dictionary is formed through the following processes: preprocessing the ICD name text in the ICD coding training data set to generate an ICD thesaurus; Organize it into a data source file; load the data source file to the long-short-term memory neural network model for training, and output it as an entity dictionary. The invention can automatically and intelligently recommend corresponding standardized ICD codes for non-standard ICD names, with high accuracy and reliability.

Description

technical field [0001] The present application relates to the field of electrical digital data processing, and in particular to an International Classification of Diseases code recommendation method, system, corresponding equipment and storage medium. Background technique [0002] Standardization and accuracy of ICD coding are very important. However, traditional assisted coding tools have existed for many years, but the way to achieve intelligence is still at a relatively shallow level. Most of the existing coding tools are based on text, and provide possible hints based on keywords, and guide the coders to obtain the final code step by step. The core of traditional auxiliary coding is string search and string matching. Due to different naming habits and limited input time when entering text, the entered names are often different from the standard ICD names. Standardization and standardization of names is done by coders. However, subject to different coders' working yea...

AI Technical Summary

Problems solved by technology

However, subject to different coders' working years, experience, personnel quality, interpretation of standards, mastery of the huge content of the coding dictionary, and restrictions on the staffing of coders, even between different coders in the same unit, the same coder in different There is also a difference in the encoding of the period

Additionally, the number of codes that coders can standardize per day is limited and misjudgments are likely

Images