Blending Knowledge in Deep Recurrent Networks for Adverse Event Prediction at Hospital 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.
Keywords
Cite
@article{arxiv.2104.04377,
title = {Blending Knowledge in Deep Recurrent Networks for Adverse Event Prediction at Hospital Discharge},
author = {Prithwish Chakraborty and James Codella and Piyush Madan and Ying Li and Hu Huang and Yoonyoung Park and Chao Yan and Ziqi Zhang and Cheng Gao and Steve Nyemba and Xu Min and Sanjib Basak and Mohamed Ghalwash and Zach Shahn and Parthasararathy Suryanarayanan and Italo Buleje and Shannon Harrer and Sarah Miller and Amol Rajmane and Colin Walsh and Jonathan Wanderer and Gigi Yuen Reed and Kenney Ng and Daby Sow and Bradley A. Malin},
journal= {arXiv preprint arXiv:2104.04377},
year = {2021}
}
Comments
Presented at the AMIA 2021 Virtual Informatics Summit