An Automatic Grading System For Electronic Medical Records With Neural Network

The automatic grading task of electronic medical records is an important issue in the cross-field of artificial intelligence and medical health.It is the first step to realize the processing of medical texts by artificial intelligence.Electronic medical records are regarded as rich resource for a wide variety of healthcare research tasks such as further understandings of patients’ health conditions,identifying their best-matching patient category and determining patients’ customized treatment plan.However,it will take a long time to train a recorder before he can write an excellent electronic medical record.In this progress grading electronic medical records is the most important part.In this paper,we present a convolutional neural network framework to grade medical records by grading note.We regard the scoring process as a text pattern classification task with mapped sentences.This framework involves two stages.The first stage is key medical concept matching between grading note and medical records.It is important to get the semantic relatedness between sentences.We use the weighted averaging word vector to get the corresponding sentence vector and then match semantic related sentences.The second stage is text pattern classification which can predict whether the key concept in grading note is correct,missing or incorrect.In this stage,we build a convolutional neural network to classify the matched text patterns.The result shows that our neural network model performs better than other traditional machine learning grading methods.Our system makes a great progress for text pattern classification accuracy and performs much better baselines on grading process.With the grading system,we have high quality electronic medical records to do predictive and exploratory tasks.It broadens the application scenarios of text classification.