Dynamic Survival Analysis with Individualized Truncated Parametric Distributions

Preston Putzel, Padhraic Smyth, Jaehong Yu, Hua Zhong
Proceedings of AAAI Spring Symposium on Survival Prediction - Algorithms, Challenges, and Applications 2021, PMLR 146:159-170, 2021.

Abstract

Dynamic survival analysis is a variant of traditional survival analysis where time-to-event predictions are updated as new information arrives about an individual over time. In this paper we propose a new approach to dynamic survival analysis based on learning a global parametric distribution, followed by individualization via truncating and renormalizing that distribution at different locations over time. We combine this approach with a likelihood-based loss that includes predictions at every time step within an individual’s history, rather than just including one term per individual. The combination of this loss and model results in an interpretable approach to dynamic survival, requiring less fine tuning than existing methods, while still achieving good predictive performance. We evaluate the approach on the problem of predicting hospital mortality for a dataset with over 6900 COVID-19 patients.

Cite this Paper

BibTeX
@InProceedings{pmlr-v146-putzel21a, title = {Dynamic Survival Analysis with Individualized Truncated Parametric Distributions}, author = {Putzel, Preston and Smyth, Padhraic and Yu, Jaehong and Zhong, Hua}, booktitle = {Proceedings of AAAI Spring Symposium on Survival Prediction - Algorithms, Challenges, and Applications 2021}, pages = {159--170}, year = {2021}, editor = {Greiner, Russell and Kumar, Neeraj and Gerds, Thomas Alexander and van der Schaar, Mihaela}, volume = {146}, series = {Proceedings of Machine Learning Research}, month = {22--24 Mar}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v146/putzel21a/putzel21a.pdf}, url = {https://proceedings.mlr.press/v146/putzel21a.html}, abstract = {Dynamic survival analysis is a variant of traditional survival analysis where time-to-event predictions are updated as new information arrives about an individual over time. In this paper we propose a new approach to dynamic survival analysis based on learning a global parametric distribution, followed by individualization via truncating and renormalizing that distribution at different locations over time. We combine this approach with a likelihood-based loss that includes predictions at every time step within an individual’s history, rather than just including one term per individual. The combination of this loss and model results in an interpretable approach to dynamic survival, requiring less fine tuning than existing methods, while still achieving good predictive performance. We evaluate the approach on the problem of predicting hospital mortality for a dataset with over 6900 COVID-19 patients.} }
Endnote
%0 Conference Paper %T Dynamic Survival Analysis with Individualized Truncated Parametric Distributions %A Preston Putzel %A Padhraic Smyth %A Jaehong Yu %A Hua Zhong %B Proceedings of AAAI Spring Symposium on Survival Prediction - Algorithms, Challenges, and Applications 2021 %C Proceedings of Machine Learning Research %D 2021 %E Russell Greiner %E Neeraj Kumar %E Thomas Alexander Gerds %E Mihaela van der Schaar %F pmlr-v146-putzel21a %I PMLR %P 159--170 %U https://proceedings.mlr.press/v146/putzel21a.html %V 146 %X Dynamic survival analysis is a variant of traditional survival analysis where time-to-event predictions are updated as new information arrives about an individual over time. In this paper we propose a new approach to dynamic survival analysis based on learning a global parametric distribution, followed by individualization via truncating and renormalizing that distribution at different locations over time. We combine this approach with a likelihood-based loss that includes predictions at every time step within an individual’s history, rather than just including one term per individual. The combination of this loss and model results in an interpretable approach to dynamic survival, requiring less fine tuning than existing methods, while still achieving good predictive performance. We evaluate the approach on the problem of predicting hospital mortality for a dataset with over 6900 COVID-19 patients.
APA
Putzel, P., Smyth, P., Yu, J. & Zhong, H.. (2021). Dynamic Survival Analysis with Individualized Truncated Parametric Distributions. Proceedings of AAAI Spring Symposium on Survival Prediction - Algorithms, Challenges, and Applications 2021, in Proceedings of Machine Learning Research 146:159-170 Available from https://proceedings.mlr.press/v146/putzel21a.html.

Related Material