Marked Point Process for Severity of Illness Assessment

Kazi T. Islam, Christian R. Shelton, Juan I. Casse, Randall Wetzel
Proceedings of the 2nd Machine Learning for Healthcare Conference, PMLR 68:255-270, 2017.

Abstract

Electronic Health Records (EHRs) consist of sparse, noisy, incomplete, heterogeneous and unevenly sampled clinical data of patients. They include physiological signals, lab test results, procedural events, clinical notes. Such data can be treated as a temporal stream of events of varied types occurring at irregularly spaced time points. We focus on modeling the temporal dependencies that arise due to the types, timings, and values of different events in such data. We model the event streams, including vital signs, laboratory results contained in two different datasets (MIMIC III — Medical Information Mart for Intensive Care clinical database — and data extracted from EHRs of patients in a tertiary pediatric intensive care unit) using a piecewise-constant conditional intensity model (PCIM), a type of marked point process. Our experiments capture meaningful temporal dependencies and show improvement in hospital mortality prediction over traditional ICU scoring systems.

Cite this Paper

BibTeX
@InProceedings{pmlr-v68-islam17a, title = {Marked Point Process for Severity of Illness Assessment}, author = {Islam, Kazi T. and Shelton, Christian R. and Casse, Juan I. and Wetzel, Randall}, booktitle = {Proceedings of the 2nd Machine Learning for Healthcare Conference}, pages = {255--270}, year = {2017}, editor = {Doshi-Velez, Finale and Fackler, Jim and Kale, David and Ranganath, Rajesh and Wallace, Byron and Wiens, Jenna}, volume = {68}, series = {Proceedings of Machine Learning Research}, month = {18--19 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v68/islam17a/islam17a.pdf}, url = {https://proceedings.mlr.press/v68/islam17a.html}, abstract = {Electronic Health Records (EHRs) consist of sparse, noisy, incomplete, heterogeneous and unevenly sampled clinical data of patients. They include physiological signals, lab test results, procedural events, clinical notes. Such data can be treated as a temporal stream of events of varied types occurring at irregularly spaced time points. We focus on modeling the temporal dependencies that arise due to the types, timings, and values of different events in such data. We model the event streams, including vital signs, laboratory results contained in two different datasets (MIMIC III — Medical Information Mart for Intensive Care clinical database — and data extracted from EHRs of patients in a tertiary pediatric intensive care unit) using a piecewise-constant conditional intensity model (PCIM), a type of marked point process. Our experiments capture meaningful temporal dependencies and show improvement in hospital mortality prediction over traditional ICU scoring systems.} }
Endnote
%0 Conference Paper %T Marked Point Process for Severity of Illness Assessment %A Kazi T. Islam %A Christian R. Shelton %A Juan I. Casse %A Randall Wetzel %B Proceedings of the 2nd Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2017 %E Finale Doshi-Velez %E Jim Fackler %E David Kale %E Rajesh Ranganath %E Byron Wallace %E Jenna Wiens %F pmlr-v68-islam17a %I PMLR %P 255--270 %U https://proceedings.mlr.press/v68/islam17a.html %V 68 %X Electronic Health Records (EHRs) consist of sparse, noisy, incomplete, heterogeneous and unevenly sampled clinical data of patients. They include physiological signals, lab test results, procedural events, clinical notes. Such data can be treated as a temporal stream of events of varied types occurring at irregularly spaced time points. We focus on modeling the temporal dependencies that arise due to the types, timings, and values of different events in such data. We model the event streams, including vital signs, laboratory results contained in two different datasets (MIMIC III — Medical Information Mart for Intensive Care clinical database — and data extracted from EHRs of patients in a tertiary pediatric intensive care unit) using a piecewise-constant conditional intensity model (PCIM), a type of marked point process. Our experiments capture meaningful temporal dependencies and show improvement in hospital mortality prediction over traditional ICU scoring systems.
APA
Islam, K.T., Shelton, C.R., Casse, J.I. & Wetzel, R.. (2017). Marked Point Process for Severity of Illness Assessment. Proceedings of the 2nd Machine Learning for Healthcare Conference, in Proceedings of Machine Learning Research 68:255-270 Available from https://proceedings.mlr.press/v68/islam17a.html.

Related Material