Boosting Deep Learning Risk Prediction with Generative Adversarial Networks for Electronic Health Records

TL;DR

This paper introduces ehrGAN, a generative adversarial network that augments limited EHR data to improve deep learning-based risk prediction in healthcare, demonstrating significant performance gains.

Contribution

The work presents a novel semi-supervised framework combining ehrGAN with CNNs to enhance risk prediction from limited EHR data, outperforming existing methods.

Findings

ehrGAN generates realistic patient data samples

The framework improves classification accuracy on healthcare datasets

Significant performance gains over state-of-the-art baselines

Abstract

The rapid growth of Electronic Health Records (EHRs), as well as the accompanied opportunities in Data-Driven Healthcare (DDH), has been attracting widespread interests and attentions. Recent progress in the design and applications of deep learning methods has shown promising results and is forcing massive changes in healthcare academia and industry, but most of these methods rely on massive labeled data. In this work, we propose a general deep learning framework which is able to boost risk prediction performance with limited EHR data. Our model takes a modified generative adversarial network namely ehrGAN, which can provide plausible labeled EHR data by mimicking real patient records, to augment the training dataset in a semi-supervised learning manner. We use this generative model together with a convolutional neural network (CNN) based prediction model to improve the onset prediction performance. Experiments on two real healthcare datasets demonstrate that our proposed framework produces realistic data samples and achieves significant improvements on classification tasks with the generated data over several stat-of-the-art baselines.