Yuan Xue
ABSTRACT
Capturing the inter-dependencies among multiple types of clinically-critical events is critical not only to accurate future event prediction, but also to better treatment planning. In this work, we propose a deep latent state-space generative model to capture the interactions among different types of correlated clinical events (e.g., kidney failure, mortality) by explicitly modeling the temporal dynamics of patients' latent states. Based on these learned patient states, we further develop a new general discrete-time formulation of the hazard rate function to estimate the survival distribution of patients with significantly improved accuracy. Extensive evaluations over real EMR data show that our proposed model compares favorably to various state-of-the-art baselines. Furthermore, our method also uncovers meaningful insights about the latent correlations among mortality and different types of organ failures.
- Martín Abadi and et al. 2015. TensorFlow: A System for Large-Scale Machine Learning.Google Scholar
- Ahmed M. Alaa and Mihaela van der Schaar. 2017. Deep Multi-task Gaussian Processes for Survival Analysis with Competing Risks. In NIPS. 2329--2337.Google Scholar
- Ahmed M. Alaa and Mihaela van der Schaar. 2019. Attentive State-Space Modeling of Disease Progression. In NeurIPS.Google Scholar
- James E. Barretta and Anthony C. C. Coolena. 2010. Gaussian process regression for survival data with competing risks.Google Scholar
- Alexis Bellot and Mihaela van der Schaar. 2018. Multitask Boosting for Survival Analysis with Competing Risks. In NeurIPS.Google Scholar
- Paidamoyo Chapfuwa, Chenyang Tao, Chunyuan Li, Courtney Page, Benjamin Goldstein, Lawrence Carin, and Ricardo Henao. 2018. Adversarial Time-to-Event Modeling. In ICML.Google Scholar
- Zhengping Che, Sanjay Purushotham, Kyunghyun Cho, David Sontag, and YanLiu. 2018. Recurrent Neural Networks for Multivariate Time Series with Missing Values. Sci. Rep.8, 1 (2018).Google Scholar
- Edward Choi, Mohammad Taha Bahadori, Andy Schuetz, Walter F Stewart, and Jimeng Sun. 2015. Doctor AI: Predicting Clinical Events via Recurrent Neural Networks. (Nov. 2015). arXiv:1511.05942Google Scholar
- D. R. Cox. 1992. Regression models and life-tables. In Breakthroughs in statistics. 527--541.Google Scholar
- Tamara Fernandez, Nicolas Rivera, and Yee Whye Teh. 2016. Gaussian Processes for Survival Analysis. In NIPS. 5021--5029.Google Scholar
- J. P. Fine and R. J. Gray. 1999. A Proportional Hazards Model for the Sub distribution of a Competing Risk. Journal of the American statistical association 94, 446(1999), 496--509.Google ScholarCross Ref
- Marco Fraccaro, Simon Kamronn, Ulrich Paquet, and Ole Winther. 2017. A Disentangled Recognition and Nonlinear Dynamics Model for Unsupervised Learning. In NIPS.Google Scholar
- Marco Fraccaro, Søren Kaae Sønderby, Ulrich Paquet, and Ole Winther. 2016. Sequential Neural Models with Stochastic Layers. In NIPS.Google ScholarDigital Library
- Marzyeh Ghassemi, Mike Wu, Michael C. Hughes, Peter Szolovits, and Finale Doshi-Velez. 2017. Predicting intervention onset in the ICU with switching state space models. American Medical Informatics Association (AMIA),.Google Scholar
- E. Giunchiglia, A. Nemchenko, and M. van der Schaar. 2018. RNN-SURV: A Deep Recurrent Model for Survival Analysis. In International Conference on Artificial Neural Networks (ICANN).Google Scholar
- Alistair E.W. Johnson, Tom J. Pollard, Lu Shen, Li wei H. Lehman, Mengling Feng,Mohammad Ghassemi, Benjamin Moody, Peter Szolovits, Leo Anthony Celi, andRoger G. Mark. 2016. MIMIC-III, A Freely Accessible Critical Care Database. Scientific Data 3 (2016). Article number: 160035.Google Scholar
- Alan E. Jones, Stephen Trzeciak, and MD Jeffrey A. Kline. 2010. The Sequential Organ Failure Assessment score for predicting outcome in patients with severe sepsis and evidence of hypoper fusion at the time of emergency department presentation.Google Scholar
- Maximilian Karl, Maximilian Soelch, Justin Bayer, and Patrick van der Smagt. 2017. Deep Variational Bayes Filters: Unsupervised Learning of State Space Models from Raw Data. In ICLR.Google Scholar
- Jared L. Katzman, Uri Shaham, Alexander Cloninger, Jonathan Bates, Tingting Jiang, and Yuval Kluger. 2018. DeepSurv: personalized treatment recommender system using a Cox proportional hazards deep neural network. BMC Medical Research Methodology 18, 1 (2018), 24.Google ScholarCross Ref
- Rahul G. Krishnan, Uri Shalit, and David Sontag. 2015. Deep Kalman Filters. CoRRabs/1511.05121 (2015).Google Scholar
- Rahul G Krishnan, Uri Shalit, and David Sontag. 2017. Structured Inference Networks for Nonlinear State Space Models. In AAAI.Google Scholar
- Changhee Lee, William R. Zame, Jinsung Yoon, and Mihaela van der Schaar. 2018. DeepHit: A Deep Learning Approach to Survival Analysis with Competing Risks. In AAAI.Google Scholar
- Zachary C Lipton, David C Kale, Charles Elkan, and Randall Wetzel. 2016. Learning to Diagnose with LSTM Recurrent Neural Networks. In International Conference on Learning Representations (ICLR).Google Scholar
- Zitao Liu and Milos Hauskrecht. 2013. Clinical Time Series Prediction with a Hierarchical Dynamical System.Artificial Intelligence in Medicine(2013), 227--237.Google Scholar
- Zitao Liu and Milos Hauskrecht. 2016. Learning Adaptive Forecasting Models from Irregularly Sampled Multivariate Clinical Data. In AAAI.Google Scholar
- Jiayu Zhou Dongxiao Zhu Lu Wang, Yan Li and Jieping Ye. 2017. Multi-task Survival Analysis. In IEEE International Conference on Data Mining.Google Scholar
- Wu M, Ghassemi M, Feng M, Celi LA, Szolovits P, and Doshi-Velez F. 2017. Understanding vasopressor intervention and weaning: risk prediction in a public heterogeneous clinical time series database. J Am Med Inform Assoc(2017),488--495.Google Scholar
- Aniruddh Raghu, Matthieu Komorowski, Leo Anthony Celi, Peter Szolovits, and Marzyeh Ghassemi. 2017. Continuous State-Space Models for Optimal Sepsis Treatment: a Deep Reinforcement Learning Approach. In Proceedings of the 2nd Machine Learning for Healthcare Conference. 147--163.Google Scholar
- Alvin Rajkomar and et al. 2018. Scalable and Accurate Deep Learning with Electronic Health Records. Digital Medicine1 (2018). Article number: 18.Google Scholar
- Rajesh Ranganath, Adler Perotte, Noémie Elhadad, and David Blei. 2016. Deep Survival Analysis. In Proceedings of Machine Learning Research. Vol. 56. 101--114.Google Scholar
- Kan Ren, Jiarui Qin, Lei Zheng, Zhengyu Yang, Weinan Zhang, Lin Qiu, and Yong Yu. 2019. Deep Recurrent Survival Analysis. In AAAI.Google Scholar
- Ying Sha and May D Wang. 2017. Interpretable Predictions of Clinical Out-comes with An Attention-based Recurrent Neural Network. In Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics.Google ScholarDigital Library
- Huan Song, Deepta Rajan, Jayaraman J Thiagarajan, and Andreas Spanias. 2018.Attend and Diagnose: Clinical Time Series Analysis Using Attention Models. InAAAI.Google Scholar
- Yuan Xue, Denny Zhou, Nan Du, Andrew M. Dai, Zhen Xu, Kun Zhang, and Claire Cui. 2019. Deep Physiological State Space Model for Clinical Forecasting. In NeurIPS Workshop on Machine Learning for Health.Google Scholar
- Jieping Ye Yan Li, Jie Wang and Chandan K. Reddy. 2016. A Multi-Task Learning Formulation for Survival Analysis. In 22nd ACM SIGKDD.Google Scholar
- Quan Zhang and Mingyuan Zhou. 2018. Nonparametric Bayesian Lomax delegate racing for survival analysis with competing risks. In NeurIPS.Google Scholar
Index Terms
Deep State-Space Generative Model For Correlated Time-to-Event Predictions
Recommendations
Multi-view learning via multiple graph regularized generative model
Topic models, such as probabilistic latent semantic analysis (PLSA) and latent Dirichlet allocation (LDA), have shown impressive success in many fields. Recently, multi-view learning via probabilistic latent semantic analysis (MVPLSA), is also designed ...
Bayes linear kinematics in a dynamic survival model
Bayes linear kinematics and Bayes linear Bayes graphical models provide an extension of Bayes linear methods so that full conditional updates may be combined with Bayes linear belief adjustment. In this paper we investigate the application of this ...
SrVARM: State Regularized Vector Autoregressive Model for Joint Learning of Hidden State Transitions and State-Dependent Inter-Variable Dependencies from Multi-variate Time Series
WWW '21: Proceedings of the Web Conference 2021Many applications, e.g., healthcare, education, call for effective methods methods for constructing predictive models from high dimensional time series data where the relationship between variables can be complex and vary over time. In such settings, ...
Comments