Neural Pharmacodynamic State Space Modeling

Zeshan M Hussain, Rahul G. Krishnan, David Sontag
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:4500-4510, 2021.

Abstract

Modeling the time-series of high-dimensional, longitudinal data is important for predicting patient disease progression. However, existing neural network based approaches that learn representations of patient state, while very flexible, are susceptible to overfitting. We propose a deep generative model that makes use of a novel attention-based neural architecture inspired by the physics of how treatments affect disease state. The result is a scalable and accurate model of high-dimensional patient biomarkers as they vary over time. Our proposed model yields significant improvements in generalization and, on real-world clinical data, provides interpretable insights into the dynamics of cancer progression.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-hussain21a, title = {Neural Pharmacodynamic State Space Modeling}, author = {Hussain, Zeshan M and Krishnan, Rahul G. and Sontag, David}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {4500--4510}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/hussain21a/hussain21a.pdf}, url = {https://proceedings.mlr.press/v139/hussain21a.html}, abstract = {Modeling the time-series of high-dimensional, longitudinal data is important for predicting patient disease progression. However, existing neural network based approaches that learn representations of patient state, while very flexible, are susceptible to overfitting. We propose a deep generative model that makes use of a novel attention-based neural architecture inspired by the physics of how treatments affect disease state. The result is a scalable and accurate model of high-dimensional patient biomarkers as they vary over time. Our proposed model yields significant improvements in generalization and, on real-world clinical data, provides interpretable insights into the dynamics of cancer progression.} }
Endnote
%0 Conference Paper %T Neural Pharmacodynamic State Space Modeling %A Zeshan M Hussain %A Rahul G. Krishnan %A David Sontag %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-hussain21a %I PMLR %P 4500--4510 %U https://proceedings.mlr.press/v139/hussain21a.html %V 139 %X Modeling the time-series of high-dimensional, longitudinal data is important for predicting patient disease progression. However, existing neural network based approaches that learn representations of patient state, while very flexible, are susceptible to overfitting. We propose a deep generative model that makes use of a novel attention-based neural architecture inspired by the physics of how treatments affect disease state. The result is a scalable and accurate model of high-dimensional patient biomarkers as they vary over time. Our proposed model yields significant improvements in generalization and, on real-world clinical data, provides interpretable insights into the dynamics of cancer progression.
APA
Hussain, Z.M., Krishnan, R.G. & Sontag, D.. (2021). Neural Pharmacodynamic State Space Modeling. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:4500-4510 Available from https://proceedings.mlr.press/v139/hussain21a.html.

Related Material