Abstract
Recurrent neural networks (RNNs) offer state-of-the-art (SOTA) performance on a variety of important prediction tasks with clinical time series data. However, meaningful translation to actionable decisions requires capability to quantify confidence in the predictions, and to address the inherent ambiguities in the data and the associated modeling process. We propose a Bayesian LSTM framework using Bayes by Backprop to characterize both modelling (epistemic) and data related (aleatoric) uncertainties in prediction tasks for clinical time series data. We evaluate our approach on mortality prediction tasks with two public Intensive Care Unit (ICU) data sets, namely, the MIMIC-III and the PhysioNet 2012 collections. We demonstrate the potential for improved performance over SOTA methods, and characterize aleatoric uncertainty in the setting of noisy features. Importantly, we demonstrate how our uncertainty estimates could be used in realistic prediction scenarios to better interpret the reliability of the data and the model predictions, and improve relevance for decision support.
S. Ramasamy and P. Krishnaswamy - Equal Contribution
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
Adams, J.R., Marlin, B.M.: Learning time series detection models from temporally imprecise labels. In: Proceedings of Machine Learning Research, vol. 54 (2017)
Baytas, I.M., Xiao, C., Zhang, X., Wang, F., Jain, A.K., Zhou, J.: Patient subtyping via time-aware LSTM networks. In: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 65–74. ACM (2017)
C. Blundell, J. Cornebise, K.K., Wierstra, D.: Weight uncertainty in neural networks. arXiv preprint arXiv:1505.05424 (2015)
Caballero Barajas, K.L., Akella, R.: Dynamically modeling patient’s health state from electronic medical records: a time series approach. In: Proceedings of the 21st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, vol. 31, pp. 69–78 (2015)
Challen, R., et al.: Artificial intelligence, bias and clinical safety. BMJ Qual. Saf. 28(3), 231–237 (2019)
Choi, E., et al: Doctor AI: predicting clinical events via recurrent neural networks. In: Machine Learning for Healthcare Conference, pp. 301–318 (2016)
Depeweg, S., Miguel, J., Hernández-Lobato, Doshi-Velez, F., Udluft, S.: Decomposition of uncertainty in Bayesian deep learning for efficient and risk-sensitive learning. arXiv preprint arXiv:1710.07283 (2018)
Fortunato, M., Blundell, C., Vinyals, O.: Bayesian recurrent neural networks. arXiv:1704.02798v4 (2019)
Gal, Y., Ghahramani, Z.: A theoretically grounded application of dropout in recurrent neural networks. In: Neural Information Processing Systems (2016)
Ghassemi, M., et al: A multivariate timeseries modeling approach to severity of illness assessment and forecasting in ICU with sparse, heterogeneous clinical data. In: Association for the Advancement of Artificial Intelligence, pp. 446–453 (2015)
Ghassemi, M., et al: Opportunities in machine learning for healthcare. arXiv:1806.00388v3 (2019)
Graves, A.: Practical variational inference for neural networks, pp. 2348–2356 (2011)
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)
Johnson, A.E., Pollard, T.J., Shen, L., Lehman, L.W., Feng, M.M., Ghassemi, M., Moody, B., Szolovits, P., Celi, L.A., Mark, R.G.: MIMIC-III, a freely accessible critical care database. Sci. Data 3, 160035 (2016)
Kendall, A., Gal, Y.: What uncertainties do we need in Bayesian deep learning for computer vision. In: Neural Information Processing Systems, vol. 30 (2017)
Lipton, Z.C., Wetzel, R., et al: Learning to diagnose with LSTM recurrent neural networks. In: International Conference on Learning Representations (2016)
MacKay, D.J.C.: A practical Bayesian framework for backpropagation networks. Neural Comput. 4(3), 448–472 (1992)
Neal, R.: Bayesian Learning for Neural Networks, vol. 118. Springer, New York (2012)
PhysioBank, P.: PhysioNet: components of a new research resource for complex physiologic signals. Circulation 101(23), e215–e220 (2000)
Quinn, J.A., et al.: Factorial switching linear dynamical systems applied to physiological condition monitoring. IEEE Trans. Pattern Anal. Mach. Intell. 31, 1537–1551 (2009)
van der Wasthuizen, J., Lasenby, J.: Bayesian LSTMs in medicine. arXiv:1706.01242 (2017)
Xiao, Y., Wang, W.Y.: Quantifying uncertainties in natural language processing tasks. In: Association for the Advancement of Artificial Intelligence, pp. 1019–1027 (2019)
Acknowledgments
Research efforts were supported by funding and infrastructure from A*STAR, Singapore (Grant Nos. SSF A1818g0044 and IAF H19/01/a0/023). The authors would also like to acknowledge inputs from Wu Jiewen and Ivan Ho Mien on the clinical datasets. We also thank Vijay Chandrasekhar for his support, and are grateful for the constructive feedback from the anonymous reviewers.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Guo, Y., Liu, Z., Ramasamy, S., Krishnaswamy, P. (2021). Uncertainty Characterization for Predictive Analytics with Clinical Time Series Data. In: Shaban-Nejad, A., Michalowski, M., Buckeridge, D.L. (eds) Explainable AI in Healthcare and Medicine. Studies in Computational Intelligence, vol 914. Springer, Cham. https://doi.org/10.1007/978-3-030-53352-6_7
Download citation
DOI: https://doi.org/10.1007/978-3-030-53352-6_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-53351-9
Online ISBN: 978-3-030-53352-6
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)