Characterizing sub-cohorts via data normalization and representation learning

Everett Rush; Ozgur Ozmen; Kathryn Knight; Byung Park; Clifton Baker; Makoto Jones; Merry Ward; Jonathan Nebeker

doi:10.1109/CBMS49503.2020.00040

Characterizing sub-cohorts via data normalization and representation learning

Everett Rush, Ozgur Ozmen, Kathryn Knight, Byung Park, Clifton Baker, Makoto Jones, Merry Ward, Jonathan Nebeker

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

The process of identifying a cohort of interest is a very challenging task. It requires manually inspecting many patient records of complex structure that might include medical coding errors and missing data. This paper presents a computational pipeline for refining the process of cohort selection based on medical concepts recorded in the electronic health records (EHRs). The pipeline extracts EHR data for a given cohort and normalizes this data using standard vocabularies. Then a stacked denoising autoencoder is used to embed the normalized patient vectors in a low dimensional space, where the patients are subsequently clustered into sub-cohorts. The goal is to represent the cohort in a standard format and abstract variants of sub-populations. As a use-case, we applied the pipeline to 1.8 million Veterans diagnosed with major depressive disorder (MDD), and identified four meaningful sub-cohorts using the features learned by the autoencoder. Then, each sub-cohort was explored using a set of keywords for interpretation.

Original language	English
Title of host publication	Proceedings - 2020 IEEE 33rd International Symposium on Computer-Based Medical Systems, CBMS 2020
Editors	Alba Garcia Seco de Herrera, Alejandro Rodriguez Gonzalez, KC Santosh, Zelalem Temesgen, Bridget Kane, Paolo Soda
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	173-176
Number of pages	4
ISBN (Electronic)	9781728194295
DOIs	https://doi.org/10.1109/CBMS49503.2020.00040
State	Published - Jul 2020
Event	33rd IEEE International Symposium on Computer-Based Medical Systems, CBMS 2020 - Virtual, Online, United States Duration: Jul 28 2020 → Jul 30 2020

Publication series

Name	Proceedings - IEEE Symposium on Computer-Based Medical Systems
Volume	2020-July
ISSN (Print)	1063-7125

Conference

Conference	33rd IEEE International Symposium on Computer-Based Medical Systems, CBMS 2020
Country/Territory	United States
City	Virtual, Online
Period	07/28/20 → 07/30/20

Funding

ACKNOWLEDGMENT This work is sponsored by the US Department of Veterans Affairs. This research used resources from the Knowledge Discovery Infrastructure at the Oak Ridge National Laboratory, which is supported by the DOE Office of Science under contract DE-AC05-00OR22725. *This manuscript has been authored by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The US government retains and the publisher, by accepting the article for publication, acknowledges that the US government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for US government purposes. DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).

Keywords

Clustering
Cohort selection
Data normalization
Electronic health records
Representation learning
UMLS

Access to Document

10.1109/CBMS49503.2020.00040

Cite this

Rush, E., Ozmen, O., Knight, K., Park, B., Baker, C., Jones, M., Ward, M., & Nebeker, J. (2020). Characterizing sub-cohorts via data normalization and representation learning. In A. G. S. de Herrera, A. Rodriguez Gonzalez, KC. Santosh, Z. Temesgen, B. Kane, & P. Soda (Eds.), Proceedings - 2020 IEEE 33rd International Symposium on Computer-Based Medical Systems, CBMS 2020 (pp. 173-176). Article 9183159 (Proceedings - IEEE Symposium on Computer-Based Medical Systems; Vol. 2020-July). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CBMS49503.2020.00040

Rush, Everett ; Ozmen, Ozgur ; Knight, Kathryn et al. / Characterizing sub-cohorts via data normalization and representation learning. Proceedings - 2020 IEEE 33rd International Symposium on Computer-Based Medical Systems, CBMS 2020. editor / Alba Garcia Seco de Herrera ; Alejandro Rodriguez Gonzalez ; KC Santosh ; Zelalem Temesgen ; Bridget Kane ; Paolo Soda. Institute of Electrical and Electronics Engineers Inc., 2020. pp. 173-176 (Proceedings - IEEE Symposium on Computer-Based Medical Systems).

@inproceedings{1053d10ea054431b810f065c47b0a269,

title = "Characterizing sub-cohorts via data normalization and representation learning",

abstract = "The process of identifying a cohort of interest is a very challenging task. It requires manually inspecting many patient records of complex structure that might include medical coding errors and missing data. This paper presents a computational pipeline for refining the process of cohort selection based on medical concepts recorded in the electronic health records (EHRs). The pipeline extracts EHR data for a given cohort and normalizes this data using standard vocabularies. Then a stacked denoising autoencoder is used to embed the normalized patient vectors in a low dimensional space, where the patients are subsequently clustered into sub-cohorts. The goal is to represent the cohort in a standard format and abstract variants of sub-populations. As a use-case, we applied the pipeline to 1.8 million Veterans diagnosed with major depressive disorder (MDD), and identified four meaningful sub-cohorts using the features learned by the autoencoder. Then, each sub-cohort was explored using a set of keywords for interpretation.",

keywords = "Clustering, Cohort selection, Data normalization, Electronic health records, Representation learning, UMLS",

author = "Everett Rush and Ozgur Ozmen and Kathryn Knight and Byung Park and Clifton Baker and Makoto Jones and Merry Ward and Jonathan Nebeker",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 33rd IEEE International Symposium on Computer-Based Medical Systems, CBMS 2020 ; Conference date: 28-07-2020 Through 30-07-2020",

year = "2020",

month = jul,

doi = "10.1109/CBMS49503.2020.00040",

language = "English",

series = "Proceedings - IEEE Symposium on Computer-Based Medical Systems",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "173--176",

editor = "{de Herrera}, {Alba Garcia Seco} and {Rodriguez Gonzalez}, Alejandro and KC Santosh and Zelalem Temesgen and Bridget Kane and Paolo Soda",

booktitle = "Proceedings - 2020 IEEE 33rd International Symposium on Computer-Based Medical Systems, CBMS 2020",

}

Rush, E, Ozmen, O, Knight, K , Park, B, Baker, C, Jones, M, Ward, M & Nebeker, J 2020, Characterizing sub-cohorts via data normalization and representation learning. in AGS de Herrera, A Rodriguez Gonzalez, KC Santosh, Z Temesgen, B Kane & P Soda (eds), Proceedings - 2020 IEEE 33rd International Symposium on Computer-Based Medical Systems, CBMS 2020., 9183159, Proceedings - IEEE Symposium on Computer-Based Medical Systems, vol. 2020-July, Institute of Electrical and Electronics Engineers Inc., pp. 173-176, 33rd IEEE International Symposium on Computer-Based Medical Systems, CBMS 2020, Virtual, Online, United States, 07/28/20. https://doi.org/10.1109/CBMS49503.2020.00040

Characterizing sub-cohorts via data normalization and representation learning. / Rush, Everett; Ozmen, Ozgur; Knight, Kathryn et al.
Proceedings - 2020 IEEE 33rd International Symposium on Computer-Based Medical Systems, CBMS 2020. ed. / Alba Garcia Seco de Herrera; Alejandro Rodriguez Gonzalez; KC Santosh; Zelalem Temesgen; Bridget Kane; Paolo Soda. Institute of Electrical and Electronics Engineers Inc., 2020. p. 173-176 9183159 (Proceedings - IEEE Symposium on Computer-Based Medical Systems; Vol. 2020-July).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Characterizing sub-cohorts via data normalization and representation learning

AU - Rush, Everett

AU - Ozmen, Ozgur

AU - Knight, Kathryn

AU - Park, Byung

AU - Baker, Clifton

AU - Jones, Makoto

AU - Ward, Merry

AU - Nebeker, Jonathan

PY - 2020/7

Y1 - 2020/7

N2 - The process of identifying a cohort of interest is a very challenging task. It requires manually inspecting many patient records of complex structure that might include medical coding errors and missing data. This paper presents a computational pipeline for refining the process of cohort selection based on medical concepts recorded in the electronic health records (EHRs). The pipeline extracts EHR data for a given cohort and normalizes this data using standard vocabularies. Then a stacked denoising autoencoder is used to embed the normalized patient vectors in a low dimensional space, where the patients are subsequently clustered into sub-cohorts. The goal is to represent the cohort in a standard format and abstract variants of sub-populations. As a use-case, we applied the pipeline to 1.8 million Veterans diagnosed with major depressive disorder (MDD), and identified four meaningful sub-cohorts using the features learned by the autoencoder. Then, each sub-cohort was explored using a set of keywords for interpretation.

AB - The process of identifying a cohort of interest is a very challenging task. It requires manually inspecting many patient records of complex structure that might include medical coding errors and missing data. This paper presents a computational pipeline for refining the process of cohort selection based on medical concepts recorded in the electronic health records (EHRs). The pipeline extracts EHR data for a given cohort and normalizes this data using standard vocabularies. Then a stacked denoising autoencoder is used to embed the normalized patient vectors in a low dimensional space, where the patients are subsequently clustered into sub-cohorts. The goal is to represent the cohort in a standard format and abstract variants of sub-populations. As a use-case, we applied the pipeline to 1.8 million Veterans diagnosed with major depressive disorder (MDD), and identified four meaningful sub-cohorts using the features learned by the autoencoder. Then, each sub-cohort was explored using a set of keywords for interpretation.

KW - Clustering

KW - Cohort selection

KW - Data normalization

KW - Electronic health records

KW - Representation learning

KW - UMLS

UR - http://www.scopus.com/inward/record.url?scp=85091133752&partnerID=8YFLogxK

U2 - 10.1109/CBMS49503.2020.00040

DO - 10.1109/CBMS49503.2020.00040

M3 - Conference contribution

AN - SCOPUS:85091133752

T3 - Proceedings - IEEE Symposium on Computer-Based Medical Systems

SP - 173

EP - 176

BT - Proceedings - 2020 IEEE 33rd International Symposium on Computer-Based Medical Systems, CBMS 2020

A2 - de Herrera, Alba Garcia Seco

A2 - Rodriguez Gonzalez, Alejandro

A2 - Santosh, KC

A2 - Temesgen, Zelalem

A2 - Kane, Bridget

A2 - Soda, Paolo

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 33rd IEEE International Symposium on Computer-Based Medical Systems, CBMS 2020

Y2 - 28 July 2020 through 30 July 2020

ER -

Rush E, Ozmen O, Knight K , Park B, Baker C, Jones M et al. Characterizing sub-cohorts via data normalization and representation learning. In de Herrera AGS, Rodriguez Gonzalez A, Santosh KC, Temesgen Z, Kane B, Soda P, editors, Proceedings - 2020 IEEE 33rd International Symposium on Computer-Based Medical Systems, CBMS 2020. Institute of Electrical and Electronics Engineers Inc. 2020. p. 173-176. 9183159. (Proceedings - IEEE Symposium on Computer-Based Medical Systems). doi: 10.1109/CBMS49503.2020.00040

Characterizing sub-cohorts via data normalization and representation learning

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this