Deep Learning on Electronic Health Records to Improve Disease Coding Accuracy

Abstract

Characterization of a patient's clinical phenotype is central to biomedical informatics. ICD codes, assigned to inpatient encounters by coders, is important for population health and cohort discovery when clinical information is limited. While ICD codes are assigned to patients by professionals trained and certified in coding there is substantial variability in coding. We present a methodology that uses deep learning methods to model coder decision making and that predicts ICD codes. Our approach predicts codes based on demographics, lab results, and medications, as well as codes from previous encounters. We are able to predict existing codes with high accuracy for all three of the test cases we investigated: diabetes, acute renal failure, and chronic kidney disease. We employed a panel of clinicians, in a blinded manner, to assess ground truth and compared the predictions of coders, model and clinicians. When disparities between the model prediction and coder assigned codes were reviewed, our model outperformed coder assigned ICD codes.

Figure 1.

(A) F1 comparison of different learning methods across three different diseases. Deep Learning (DL), Logistic Regression (LR) and Random Forest (RF) are shown. (B) Performance of deep learning prediction in other data sets, for facilities with identifier 143, 67, or the multi-10 dataset consisting of 10 different facilities.

Figure 2.

Using expert review as ground truth, data shows the success of model predictions in cases where they disagree with existing codes. Model accuracy is shown as median (dot) and the 95% confidence interval (bar) using a beta-binomial model with Jeffreys prior. CKD - Chronic Kidney Disease, ARF - Acute Renal Failure.