Abstract
Synthetic electronic health records (EHR) can facilitate effective use of clinical data in software development, medical education, and medical research without the concerns of data privacy. We propose a novel Generative Adversarial Network (GAN) approach, called Longitudinal GAN (LongGAN), that can generate synthetic longitudinal EHR data. LongGAN employs a recurrent autoencoder and the Wasserstein GAN Gradient Penalty (WGAN-GP) architecture with conditional inputs. We evaluate LongGAN with the task of generating training data for machine/deep learning methods. Our experiments show that predictive models trained with synthetic data from LongGAN achieve comparable performance to those trained with real data. Moreover, these models have up to 0.27 higher AUROC and up to 0.21 higher AUPRC values than models trained with synthetic data from RCGAN and TimeGAN, the two most relevant methods for longitudinal data generation. We also demonstrate that LongGAN is able to preserve patient privacy in a given attribute disclosure attack setting.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Rothstein, M.A.: Is deidentification sufficient to protect health privacy in research? Am J Bioeth. 10(9), 3–11 (2010)
Foraker, R.E., Yu, S.C., Gupta, A., Michelson, A.P., Pineda Soto, J.A., Colvin, R., et al.: Spot the difference: Comparing results of analyses from real patient data and synthetic derivatives. JAMIA Open. 3(4), 557–566 (2020)
Benaim, A.R., et al.: Analyzing medical research results based on synthetic data and their relation to real data results: Systematic comparison from five observational studies. JMIR Med. Inform. 8(2), e16492 (2020)
Guo, A., Foraker, R.E., MacGregor, R.M., Masood, F.M., Cupps, B.P., Pasque, M.K.: The use of synthetic electronic health record data and deep learning to improve timing of high-risk heart failure surgical intervention by predicting proximity to catastrophic decompensation. Front. Digit. Health 44 (2020)
Che, Z., Cheng, Y., Zhai, S., Sun, Z., Liu, Y.: Boosting deep learning risk prediction with generative adversarial networks for electronic health records. In: 2017 IEEE International Conference on Data Mining (ICDM), pp. 787–92 (2017)
Walonoski, J.A., Kramer, M., Nichols, J., Quina, A., Moesel, C., Hall, D., et al.: Synthea: an approach, method, and software mechanism for generating synthetic patients and the synthetic electronic health care record. J. Am. Med. Inf. Assoc. JAMIA. 25, 230–238 (2018)
Dube, K., Gallagher, T.: Approach and Method for Generating Realistic Synthetic Electronic Healthcare Records for Secondary Use. In: Gibbons, J., MacCaull, W. (eds.) FHIES 2013. LNCS, vol. 8315, pp. 69–86. Springer, Heidelberg (2014). https://doi.org/10.1007/978-3-642-53956-5_6
Goncalves, A., Ray, P., Soper, B., Stevens, J., Coyle, L., Sales, A.P.: Generation and evaluation of synthetic patient data. BMC Med. Res. Method. 20(1), 1–40 (2020)
McLachlan, S., Dube, K., Gallagher, T., Simmonds, J.A., Fenton, N.: Realistic Synthetic Data Generation: The ATEN Framework. In: Cliquet Jr., A., et al. (eds.) BIOSTEC 2018. CCIS, vol. 1024, pp. 497–523. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-29196-9_25
Pollack, A.H., Simon, T.D., Snyder, J., Pratt, W.: Creating synthetic patient data to support the design and evaluation of novel health information technology. J. Biomed. Inf. 95, 103201 (2019)
Walonoski, J., et al.: Synthe‚ novel coronavirus (covid-19) model and synthetic data set. Intell. Based Med. 1, 100007 (2020)
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9, 1735–1780 (1997)
Dong X, et al.: Identifying risk of opioid use disorder for patients taking opioid medications with deep learning. arXiv preprint arXiv:201004589 (2020)
Dong, X., et al.: Predicting opioid overdose risk of patients with opioid prescriptions using electronic health records based on temporal deep learning. J. Biomed. Inf. 116, 103725 (2021)
Rashidian, S., et al.: Detecting miscoded diabetes diagnosis codes in electronic health records for quality improvement: temporal deep learning approach. JMIR Med. Inform. 8(12), e22649 (2020)
Tao, M., Tang, H., Wu, S., Sebe, N., Wu, F., Jing, X.: Df-gan: deep fusion generative adversarial networks for text-to-image synthesis. ArXiv. abs/2008.05865 (2020)
Clark, A., Donahue, J., Simonyan, K.: Adversarial video generation on complex datasets. arXiv: Computer Vision and Pattern Recognition (2019)
Engel, J., Agrawal, K.K., Chen, S., Gulrajani, I., Donahue, C., Roberts, A.: Gansynth: adversarial neural audio synthesis. ArXiv; abs/1902.08710 (2019)
Goodfellow, I., et al.: Generative adversarial nets. In: NIPS (2014).
Choi, E., Biswal, S., Malin, B., Duke, J., Stewart, W., Sun, J.: Generating multi-label discrete electronic health records using generative adversarial networks. ArXiv; abs/1703.06490 (2017)
Rashidian, S., et al.: SMOOTH-GAN: towards sharp and smooth synthetic ehr data generation. In: Michalowski, M., Moskovitch, R. (eds.) Artificial Intelligence in Medicine. AIME 2020. Lecture Notes in Computer Science, vol. 12299. Springer, Cham (2020)
Esteban, C., Hyland, S.L., Rätsch, G.: Real-valued (medical) time series generation with recurrent conditional GANs. ArXiv. abs/1706.02633 (2017)
Yoon, J., Jarrett, D., Schaar, M.V.D.: Time-series generative adversarial networks. In: NeurIPS (2019)
Lee, D., Yu, H., Jiang, X., Rogith, D., Gudala, M., Tejani, M., et al.: Generating sequential electronic health records using dual adversarial autoencoder. J. Am. Med. Inform. Assoc. 27(9), 1411–1419 (2020)
Jordon, J., Yoon, J., Schaar, M.V.D.: Pate-gan: generating synthetic data with differential privacy guarantees. In: ICLR (2019)
Baowaly, M.K., Lin, C., Liu, C.-L., Chen, K.-T.: Synthesizing electronic health records using improved generative adversarial networks. J. Am. Med. Inform. Assoc. 26(228), 41 (2019)
Yoon, J., Drumright, L.N., Van Der Schaar, M.: Anonymization through data synthesis using generative adversarial networks (ads-gan). IEEE J. Biomed. Health Informatics. 24(8), 2378–2388
Gulrajani, I., Ahmed, F., Arjovsky, M., Dumoulin, V., Courville, A.: Improved training of wasserstein gans. In: NIPS (2017)
Nguyen, H.D., Tran, K.P., Thomassey, S., Hamad, M.: Forecasting and anomaly detection approaches using LSTM and LSTM autoencoder techniques with the applications in supply chain management. Int. J. Inf. Manag. 57, 102282 (2021)
Chawla, A., Lee, B., Jacob, P., Fallon, S.: Bidirectional LSTM autoencoder for sequence based anomaly detection in cyber security. Int. J. Simulation: Syst., Sci. Technol. (2019)
Wong, T., Luo, Z.: Recurrent auto-encoder model for multidimensional time series representation (2018)
Mirza, M, Osindero, S.: Conditional generative adversarial nets. ArXiv. abs/1411.1784 (2014)
Al-Shawwa, B., Glynn, E., Hoffman, M.A., Ehsan, Z., Ingram, D.G.: Outpatient health care utilization for sleep disorders in the cerner health facts database. J. Clin. Sleep Med. 17(2), 203–209 (2021)
Petrick, J.L., Nguyen, T., Cook, M.B.: Temporal trends of esophageal disorders by age in the cerner health facts database. Ann. Epidemiol. 26(2), 151–4.e4 (2016)
DeShazo, J.P., Hoffman, M.: A comparison of a multistate inpatient ehr database to the hcup nationwide inpatient sample. BMC Health Services Res. 15(1), 1–8 (2015)
Hripcsak, G., Ryan, P.B., Duke, J.D., Shah, N.H., Park, R.W., Huser, V., et al.: Characterizing treatment pathways at scale using the ohdsi network. Proc Natl Acad Sci U S A. 113(27), 7329–7336 (2016)
Shukla, S.N., Marlin, B.M.: Interpolation-prediction networks for irregularly sampled time series. ArXiv ;abs/1909.07782 (2019)
Müller, R., Kornblith, S., Hinton, G.E.: When does label smoothing help? In: NeurIPS (2019)
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., et al.: Scikit-learn: Machine learning in python. J. Mach. Learn. Res. 12, 2825–2830 (2011)
Abadi, M., et al.: Tensorflow: a system for large-scale machine learning. In: OSDI (2016)
Oliphant, T.E.: Guide to NumPy (2015)
McKinney, W.: Data structures for statistical computing in python (2010)
Virtanen, P., et al.: Scipy 1.0: fundamental algorithms for scientific computing in python. Nat. Method. 17(3), 261–272 (2020).
Matwin, S., Nin, J., Sehatkar, M., Szapiro, T.: A review of attribute disclosure control. In: Navarro-Arribas G., Torra V. (eds.) Advanced Research in Data Privacy. Studies in Computational Intelligence, vol. 567. Springer, Cham (2015)
Surendra, H., MohanH, S.: A review of synthetic data generation methods for privacy preserving data publishing. Int. J. Sci. Technol. Res. 6, 95–101 (2017)
Hittmeir, M., Mayer, R., Ekelhart, A.: A baseline for attribute disclosure risk in synthetic data. In: Proceedings of the Tenth ACM Conference on Data and Application Security and Privacy (2020)
Stadler, T., Oprisanu, B., Troncoso, C.: Synthetic data - a privacy mirage. ArXiv. abs/2011.07018 (2020)
Pascanu, R., Mikolov, T., Bengio, Y.: On the difficulty of training recurrent neural networks. In: ICML (2013)
García-Laencina, P.J., Sancho-Gómez, J., Figueiras-Vidal, A.R.: Pattern classification with missing data: A review. Neural Comput. Appl. 19, 263–282 (2009)
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., et al.: Attention is all you need. ArXiv. abs/1706.03762 (2017)
Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., et al.: Roberta: a robustly optimized bert pretraining approach. ArXiv. abs/1907.11692 (2019)
Devlin, J., Chang, M.W., Lee, K., Toutanova, K.: Bert: pre-training of deep bidirectional transformers for language understanding. In: NAACL-HLT (2019)
Choi, K., Hawthorne, C., Simon, I., Dinculescu, M., Engel, J.: Encoding musical style with transformer autoencoders. In: ICML (2020)
Fang, L., Zeng, T., Liu, C.C., Bo, L., Dong, W., Chen, C.: Transformer-based conditional variational autoencoder for controllable story generation. ArXiv abs/2101.00828 (2021)
Toreini, E., et al.: Technologies for trustworthy machine learning: A survey in a socio-technical context. ArXiv. abs/2007.08911 (2020)
Dwork, C., Roth, A.: The algorithmic foundations of differential privacy. Found Trends Theor. Comput. Sci. 9, 211–407 (2014)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Sun, S. et al. (2021). Generating Longitudinal Synthetic EHR Data with Recurrent Autoencoders and Generative Adversarial Networks. In: Rezig, E.K., et al. Heterogeneous Data Management, Polystores, and Analytics for Healthcare. DMAH Poly 2021 2021. Lecture Notes in Computer Science(), vol 12921. Springer, Cham. https://doi.org/10.1007/978-3-030-93663-1_12
Download citation
DOI: https://doi.org/10.1007/978-3-030-93663-1_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-93662-4
Online ISBN: 978-3-030-93663-1
eBook Packages: Computer ScienceComputer Science (R0)