arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2204.04797cs.LG
收藏

基于条件GAN的多标签临床时间序列生成

Multi-Label Clinical Time-Series Generation via Conditional GAN

Stevens Institute of Technology(史蒂文斯理工学院);Virginia Tech(弗吉尼亚理工大学);University of North Carolina at Chapel Hill(北卡罗来纳大学教堂山分校)

Chang Lu, Chandan K. Reddy, Ping Wang, Dong Nie, Yue Ning

AI总结:

提出MTGAN模型,利用条件GAN和GRU生成多标签临床时间序列数据,通过Wasserstein距离和稳定训练策略,有效提升罕见疾病数据的生成质量。

摘要:

近年来,深度学习已在电子健康记录(EHR)相关的广泛应用中取得成功,如表示学习和临床事件预测。然而,由于隐私限制,对EHR的有限访问成为深度学习研究的瓶颈。为缓解这些问题,生成对抗网络(GAN)已被成功用于生成EHR数据。然而,高质量EHR生成仍面临挑战,包括生成时间序列EHR数据和处理不平衡的罕见疾病。在本工作中,我们提出一种多标签时间序列GAN(MTGAN)来生成EHR,并同时提高罕见疾病生成的质量。MTGAN的生成器使用带平滑条件矩阵的门控循环单元(GRU)来生成序列和罕见疾病。判别器使用Wasserstein距离,通过考虑数据和时序特征来区分真实样本与合成样本。我们还提出一种训练策略,用于计算真实数据的时序特征并稳定GAN训练。此外,我们设计了多种统计指标和预测任务来评估生成的数据。实验结果表明,合成数据的质量以及MTGAN在生成逼真的序列EHR数据方面的有效性,尤其是在罕见疾病方面。

英文摘要:

In recent years, deep learning has been successfully adopted in a wide range of applications related to electronic health records (EHRs) such as representation learning and clinical event prediction. However, due to privacy constraints, limited access to EHR becomes a bottleneck for deep learning research. To mitigate these concerns, generative adversarial networks (GANs) have been successfully used for generating EHR data. However, there are still challenges in high-quality EHR generation, including generating time-series EHR data and imbalanced uncommon diseases. In this work, we propose a Multi-label Time-series GAN (MTGAN) to generate EHR and simultaneously improve the quality of uncommon disease generation. The generator of MTGAN uses a gated recurrent unit (GRU) with a smooth conditional matrix to generate sequences and uncommon diseases. The critic gives scores using Wasserstein distance to recognize real samples from synthetic samples by considering both data and temporal features. We also propose a training strategy to calculate temporal features for real data and stabilize GAN training. Furthermore, we design multiple statistical metrics and prediction tasks to evaluate the generated data. Experimental results demonstrate the quality of the synthetic data and the effectiveness of MTGAN in generating realistic sequential EHR data, especially for uncommon diseases.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑