Disease prediction device and equipment, and symptom information processing method, device and equipment

Abstract

The invention discloses a disease prediction device and equipment, and a symptom information processing method, device and equipment. The disease prediction device acquires a current symptom feature set of a user through a first acquisition unit. Through a second acquisition unit, the prior probability of the target disease and the conditional probability of each symptom feature under the target disease are acquired from the medical knowledge graph. Through a first calculating unit, an evaluation value of the target disease is calculated according to the prior probability of the target disease, the conditional probability of each symptom feature under the target disease condition and the feature value of each symptom feature. The diseases of which the evaluation values are obtained are ranked according to the evaluation values through a prediction unit, and the disease the rank of which meets conditions is determined as predicted disease corresponding to the current symptom feature set of the user. Disease prediction is carried out based on the medical knowledge graph without the need of model training using annotation data, and the cost is reduced. The prior probability and the conditional probability in the medical knowledge graph are utilized to calculate the evaluation value of the target disease, so that the disease prediction accuracy is improved.

Description

technical field [0001] The present application relates to the technical field of data processing, in particular to a disease prediction device and equipment, and a symptom information processing method, device and equipment. Background technique [0002] The automatic consultation system is used to provide consultation services to users online. The user can input the symptoms that appear into the automatic consultation system, and obtain the disease that may correspond to the symptoms output by the automatic consultation system, so that the user can follow up with the doctor or treat himself according to the obtained disease. [0003] At present, the prediction of diseases in the automatic consultation system is generally realized by establishing a machine learning model or a neural network model based on a large amount of labeled data. The cost of establishing a machine learning model or a neural network model is relatively high, and the training time is relatively long. M...

AI Technical Summary

Problems solved by technology

[0003] At present, the prediction of diseases in the automatic consultation system is generally realized by establishing a machine learning model or a neural network model based on a large amount of labeled data. The cost of establishing a machine learning model or a neural network model is relatively high, and the training time is relatively long.

Moreover, the accuracy of disease prediction in the automatic consultation system is easily affected by the quality of labeled data

Images