Timezone: »

 
Dynamic Survival Transformers for Causal Inference with Electronic Health Records
Prayag Chatha · Yixin Wang · Zhenke Wu · Jeffrey Regier

Fri Dec 02 01:00 PM -- 02:00 PM (PST) @
in Learning from Time Series for Health »
Event URL: https://openreview.net/forum?id=6quJeu5gJ7 »

In medicine, researchers often seek to infer the effects of a given treatment on patients' outcomes, such as the expected time until infection. However, the standard methods for causal survival analysis make simplistic assumptions about the data-generating process and cannot capture complex interactions among patient covariates. We introduce the Dynamic Survival Transformer (DynST), a deep survival model that trains on electronic health records (EHRs). Unlike previous transformers used in survival analysis, DynST can make use of time-varying information to predict evolving survival probabilities. We derive a semi-synthetic EHR dataset from MIMIC-III to show that DynST can accurately estimate the causal effect of a treatment intervention on restricted mean survival time (RMST). We demonstrate that DynST achieves better predictive and causal estimation than two alternative models.

Author Information

Prayag Chatha (University of Michigan - Ann Arbor)
Prayag Chatha

I am a Statistics Ph.D. student at the University of Michigan, Ann Arbor. My research focuses on machine learning for causal inference and electronic health records. My side interests include music, economic history, and literature.

Yixin Wang (University of Michigan)
Zhenke Wu (University of Michigan)

Zhenke Wu’s research involves the development of statistical methods that inform health decisions made by individuals. He is particularly interested in scalable Bayesian methods that integrate multiple sources of evidence, with a focus on hierarchical latent variable modeling. We have applied our methods to estimate the etiology of childhood pneumonia, autoantibody signatures for subsetting autoimmune disease patients and to predict whether a user is engaged with mobile applications. Zhenke has developed original methods and software that are now used by investigators from research institutes such as US CDC and Johns Hopkins, as well as site investigators from developing countries, e.g., Kenya, South Africa, Gambia, Mali, Zambia, Thailand and Bangladesh. Zhenke completed a BS in Math at Fudan University in 2009 and a PhD in Biostatistics from the Johns Hopkins University in 2014 and then stayed at Hopkins for his postdoctoral training. Since 2016, Zhenke is Assistant Professor of Biostatistics, and Research Assistant Professor in Michigan Institute for Data Science (MIDAS) at University of Michigan, Ann Arbor.

Jeffrey Regier (University of Michigan)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors