Modeling "Presentness" of Electronic Health Record Data to Improve Patient State Estimation

Jacob Fauber, Christian R. Shelton
Proceedings of the 3rd Machine Learning for Healthcare Conference, PMLR 85:500-513, 2018.

Abstract

Medical data are not missing at random. The problem is more acute when the observations are over an extended period of time; any particular variable is observed at relatively few time points. We taking missing values to be the norm, and treat “presentness” (the times of observations) as additional features to augment the values observed. A joint model over both avoids the “missing at random” assumption. We use piecewise-constant conditional intensity models (PCIMs) to build a generative model of observation times and values. We demonstrate its effectiveness in reconstruction of monitor readings of patient vitals from sparse EHR data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v85-fauber18a, title = {Modeling ``Presentness'' of Electronic Health Record Data to Improve Patient State Estimation}, author = {Fauber, Jacob and Shelton, Christian R.}, booktitle = {Proceedings of the 3rd Machine Learning for Healthcare Conference}, pages = {500--513}, year = {2018}, editor = {Doshi-Velez, Finale and Fackler, Jim and Jung, Ken and Kale, David and Ranganath, Rajesh and Wallace, Byron and Wiens, Jenna}, volume = {85}, series = {Proceedings of Machine Learning Research}, month = {17--18 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v85/fauber18a/fauber18a.pdf}, url = {https://proceedings.mlr.press/v85/fauber18a.html}, abstract = {Medical data are not missing at random. The problem is more acute when the observations are over an extended period of time; any particular variable is observed at relatively few time points. We taking missing values to be the norm, and treat “presentness” (the times of observations) as additional features to augment the values observed. A joint model over both avoids the “missing at random” assumption. We use piecewise-constant conditional intensity models (PCIMs) to build a generative model of observation times and values. We demonstrate its effectiveness in reconstruction of monitor readings of patient vitals from sparse EHR data.} }
Endnote
%0 Conference Paper %T Modeling "Presentness" of Electronic Health Record Data to Improve Patient State Estimation %A Jacob Fauber %A Christian R. Shelton %B Proceedings of the 3rd Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2018 %E Finale Doshi-Velez %E Jim Fackler %E Ken Jung %E David Kale %E Rajesh Ranganath %E Byron Wallace %E Jenna Wiens %F pmlr-v85-fauber18a %I PMLR %P 500--513 %U https://proceedings.mlr.press/v85/fauber18a.html %V 85 %X Medical data are not missing at random. The problem is more acute when the observations are over an extended period of time; any particular variable is observed at relatively few time points. We taking missing values to be the norm, and treat “presentness” (the times of observations) as additional features to augment the values observed. A joint model over both avoids the “missing at random” assumption. We use piecewise-constant conditional intensity models (PCIMs) to build a generative model of observation times and values. We demonstrate its effectiveness in reconstruction of monitor readings of patient vitals from sparse EHR data.
APA
Fauber, J. & Shelton, C.R.. (2018). Modeling "Presentness" of Electronic Health Record Data to Improve Patient State Estimation. Proceedings of the 3rd Machine Learning for Healthcare Conference, in Proceedings of Machine Learning Research 85:500-513 Available from https://proceedings.mlr.press/v85/fauber18a.html.

Related Material