Highrisk Prediction from Electronic Medical Records via Deep Attention Networks
This work provides an incremental improvement in predicting high-risk vascular diseases for hypertension patients by utilizing readily available electronic medical record data, potentially reducing costs and analysis time compared to methods relying on pathological and radiological measurements.
This paper addresses the prediction of high-risk vascular diseases using only symbolic medical history data, such as ICD-10 and pharmacy codes. The proposed deep attention models, MeHPAN (R-MeHPAN and C-MeHPAN), were evaluated on 50,000 hypertension patients and demonstrated superior performance compared to standard classification models across precision, recall, f1-measure, and AUC.
Predicting highrisk vascular diseases is a significant issue in the medical domain. Most predicting methods predict the prognosis of patients from pathological and radiological measurements, which are expensive and require much time to be analyzed. Here we propose deep attention models that predict the onset of the high risky vascular disease from symbolic medical histories sequence of hypertension patients such as ICD-10 and pharmacy codes only, Medical History-based Prediction using Attention Network (MeHPAN). We demonstrate two types of attention models based on 1) bidirectional gated recurrent unit (R-MeHPAN) and 2) 1D convolutional multilayer model (C-MeHPAN). Two MeHPAN models are evaluated on approximately 50,000 hypertension patients with respect to precision, recall, f1-measure and area under the curve (AUC). Experimental results show that our MeHPAN methods outperform standard classification models. Comparing two MeHPANs, R-MeHPAN provides more better discriminative capability with respect to all metrics while C-MeHPAN presents much shorter training time with competitive accuracy.