Blending Knowledge in Deep Recurrent Networks for Adverse Event Prediction at Hospital Discharge

TL;DR

This paper proposes a novel deep learning architecture that integrates domain knowledge with recurrent neural networks to improve prediction of adverse events at hospital discharge, addressing data sparsity issues.

Contribution

It introduces a blended model combining self-attention RNN representations with clinical features, enhancing predictive performance over traditional methods.

Findings

Blended model outperforms standard machine learning approaches.

Deep learning benefits from incorporating domain knowledge.

Model effectively predicts adverse events at discharge.

Abstract

Deep learning architectures have an extremely high-capacity for modeling complex data in a wide variety of domains. However, these architectures have been limited in their ability to support complex prediction problems using insurance claims data, such as readmission at 30 days, mainly due to data sparsity issue. Consequently, classical machine learning methods, especially those that embed domain knowledge in handcrafted features, are often on par with, and sometimes outperform, deep learning approaches. In this paper, we illustrate how the potential of deep learning can be achieved by blending domain knowledge within deep learning architectures to predict adverse events at hospital discharge, including readmissions. More specifically, we introduce a learning architecture that fuses a representation of patient data computed by a self-attention based recurrent neural network, with clinically relevant features. We conduct extensive experiments on a large claims dataset and show that the blended method outperforms the standard machine learning approaches.