System for assisting disease reasoning and storage medium

Abstract

The present invention relates to a system for assisting disease reasoning, comprising: a storage unit which stores a knowledge graph and a multi-graph; the acquisition unit is used for acquiring the chief complaint of the user; the processing unit comprises a medical concept linking module which links the combined relation tuple entity of the chief complaint to the corresponding medical concept and obtains features in the chief complaint; a diagnosis conditional probability calculation module that recall diagnosis regarding symptoms based at least on positive and negative symptoms and calculates a diagnosis conditional probability for diagnosing relative symptoms; the user chief complaint semantic calculation module is used for calculating the semantic similarity between the chief complaint in the training corpus and the chief complaint of the user by utilizing a BERT model, screening candidate chief complaints and filtering the candidate chief complaints according to the recalled diagnosis; and a sorting algorithm module for calculating and sorting predicted values for all the determined diagnoses based on a GBDT algorithm; and a diagnosis determination module which outputs a sorting result of all diagnoses. The invention further relates to a storage medium for storing instructions for realizing the functions.

Description

technical field [0001] The invention relates to a system and a storage medium for assisting disease reasoning. Background technology [0002] The overall design of the invention can be applied to the fields of artificial intelligence, natural language processing, knowledge atlas and machine learning. [0003] Before the user visits, according to the user's main complaint description, the intelligent inquiry system can predict possible diseases. At present, intelligent inquiry systems mostly predict diseases through user complaints or multi classification algorithms combined with knowledge map. [0004] However, the above intelligent interrogation has obvious certainty. On the one hand, the intelligent interrogation system makes multi classification prediction on the Internet interrogation data, and most of these predictions are the evaluation of homologous corpus. However, in practical application, the intelligent interrogation system needs to face a variety of environments, and ...

AI Technical Summary

Problems solved by technology

[0004] However, there is an obvious certainty in the intelligent consultation mentioned above

On the one hand, the intelligent consultation system makes multi-category predictions on Internet consultation data, and these predictions are mostly homologous corpus evaluations. However, in practical applications, the intelligent consultation system needs to face various environments, and the differences in user expectations large, difficult to commercialize

On the other hand, the quality of Internet consultation data relied on by the intelligent consultation system is relatively high, and a single algorithm cannot achieve relatively accurate and effective expected results in actual products.

Images