Cardiovascular disease risk predicting method based on electronic medical record

Abstract

The invention discloses a cardiovascular disease risk predicting method based on an electronic medical record. The method comprises the steps of acquiring electronic medical record data and dividing the electronic medical record data into a training set and a testing set; arranging the electronic medical record data for forming a training set sequence and a testing set sequence; inputting the training set sequence into the embedding layer of the predicting model for generating an embedding vector; performing representation learning on the embedding vector by means of an LSTM module based on attention mechanism for obtaining a representation vector; splicing the representation vector and obtaining a preliminary cardiovascular disease risk predicting model through softmax layer prediction; performing testing and correcting on the preliminary cardiovascular disease for obtaining a final cardiovascular disease risk predicting model; and performing cardiovascular disease risk prediction onthe to-be-predicted patient by means of the final cardiovascular disease risk predicting model. The cardiovascular disease risk predicting method can efficiently and comprehensively capture characteristic information of the electronic medical record data and realizes more accurate disease risk prediction and higher model accuracy.

Description

technical field [0001] The present invention specifically relates to a cardiovascular disease risk prediction method based on electronic medical records. Background technique [0002] With the development of economy and technology and the improvement of people's living standards, cardiovascular diseases have gradually appeared widely. Cardiovascular disease is a common chronic disease that seriously threatens human health, ranking first among the total causes of death among urban and rural residents. Accurately predicting the risk of cardiovascular disease is of great significance to prevent the occurrence of cardiovascular disease. Although clinically angiography can accurately diagnose cardiovascular disease, angiography is not only expensive but also invasive to the body. In addition, electrocardiograms and some scoring indexes are commonly used clinically to predict cardiovascular risk, but these methods require doctors or practitioners to have rich theoretical knowled...

AI Technical Summary

Problems solved by technology

Although the model achieves effective risk prediction, it ignores the association between diagnosis and drug treatment, so the predictive performance suffers

Images