Abstract:

BACKGROUND: There are a large number of information in the electronic medical records which can assist medical diagnosis and decision-making;
OBJECTIVE: To perform data mining in electronic medical records based on BP neural network.
METHODS: BP artificial neural network theory and algorithms are analyzed, and there are six steps to build a BP artificial neural network model, including the training data set identification, data preparation, network model, data mining, and evaluating and predicting the results obtained from the BP network. The related applications of the BP artificial neural network in the electronic medical records were analyzed.
RESULTS AND CONCLUSION: BP artificial neural network can be used to analyze and predict the electronic medical records to find the risk factors. BP artificial neural network data analysis has practical application value in the electronic medical records system.