Medical record-based disease abnormity early warning method and device, and medium

Abstract

The invention discloses a medical record-based disease abnormity early warning method and device and a medium, and the method comprises the steps: obtaining the current medical record information, and coding the disease representation in the medical record information, so as to obtain a structured first medical record vector; performing dimension compression on the first medical record vector through a pre-trained medical record vectorization model to obtain a low-dimension second medical record vector; through a pre-trained disease trend prediction model, taking the second medical record vector as input, performing anomaly prediction on a specified disease occurring in the current medical record information, and outputting an anomaly expectation corresponding to the current medical record information according to a prediction result; and if the deviation level between the output result obtained through the disease trend prediction model and the abnormal expectation of the medical record information of the specified individual or the specified group reaches a preset level, determining that the specified individual or the specified group is abnormal.

Description

technical field [0001] The present application relates to the field of early warning of abnormalities, in particular to a method, equipment and medium for early warning of abnormal diseases based on medical records. Background technique [0002] With the continuous emergence of new diseases and the resurgence of existing diseases, the existing disease monitoring methods can no longer meet the requirements of social emergency response. In order to ensure timely and accurate prediction of disease abnormalities and take effective prevention and control measures, disease early warning is generally carried out through data processing and data mining of patient's disease information and medical treatment data. However, this method requires a large amount of data The pre-processing operation has a large workload and low processing efficiency, and the early warning effect is not good. Contents of the invention [0003] In order to solve the above problems, this application propos...

AI Technical Summary

Problems solved by technology

[0002] With the continuous emergence of new diseases and the resurgence of existing diseases, the existing disease monitoring methods can no longer meet the requirements of social emergency response

In order to ensure timely and accurate prediction of disease abnormalities and take effective prevention and control measures, disease early warning is generally carried out through data processing and data mining of patient's disease information and medical treatment data. However, this method requires a large amount of data Pre-processing operation, heavy workload and low processing efficiency, poor early warning effect

Images