Modeling the Uncertainty in Electronic Health Records: a Bayesian Deep Learning Approach
This work addresses trust issues in healthcare AI by providing uncertainty estimates, but it is incremental as it applies an existing Bayesian method to EHR data.
The paper tackles the problem of lack of transparency and trustworthiness in deep learning models for Electronic Health Records by proposing a Bayesian Neural Network to predict uncertainty from data noise, showing that high uncertainty harms performance and identifying patients for timely intervention to improve accuracy.
Deep learning models have exhibited superior performance in predictive tasks with the explosively increasing Electronic Health Records (EHR). However, due to the lack of transparency, behaviors of deep learning models are difficult to interpret. Without trustworthiness, deep learning models will not be able to assist in the real-world decision-making process of healthcare issues. We propose a deep learning model based on Bayesian Neural Networks (BNN) to predict uncertainty induced by data noise. The uncertainty is introduced to provide model predictions with an extra level of confidence. Our experiments verify that instances with high uncertainty are harmful to model performance. Moreover, by investigating the distributions of model prediction and uncertainty, we show that it is possible to identify a group of patients for timely intervention, such that decreasing data noise will benefit more on the prediction accuracy for these patients.