Unsupervised probabilistic models for sequential Electronic Health Records.
EHR data
Mixture modeling
Subgroup analysis
Unsupervised learning
Journal
Journal of biomedical informatics
ISSN: 1532-0480
Titre abrégé: J Biomed Inform
Pays: United States
ID NLM: 100970413
Informations de publication
Date de publication:
10 2022
10 2022
Historique:
received:
14
04
2022
revised:
23
06
2022
accepted:
11
08
2022
pubmed:
30
8
2022
medline:
14
10
2022
entrez:
29
8
2022
Statut:
ppublish
Résumé
We develop an unsupervised probabilistic model for heterogeneous Electronic Health Record (EHR) data. Utilizing a mixture model formulation, our approach directly models sequences of arbitrary length, such as medications and laboratory results. This allows for subgrouping and incorporation of the dynamics underlying heterogeneous data types. The model consists of a layered set of latent variables that encode underlying structure in the data. These variables represent subject subgroups at the top layer, and unobserved states for sequences in the second layer. We train this model on episodic data from subjects receiving medical care in the Kaiser Permanente Northern California integrated healthcare delivery system. The resulting properties of the trained model generate novel insight from these complex and multifaceted data. In addition, we show how the model can be used to analyze sequences that contribute to assessment of mortality likelihood.
Identifiants
pubmed: 36038064
pii: S1532-0464(22)00175-7
doi: 10.1016/j.jbi.2022.104163
pmc: PMC10588733
mid: NIHMS1936178
pii:
doi:
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Research Support, U.S. Gov't, Non-P.H.S.
Langues
eng
Sous-ensembles de citation
IM
Pagination
104163Subventions
Organisme : NIGMS NIH HHS
ID : R35 GM128672
Pays : United States
Informations de copyright
Copyright © 2022 Elsevier Inc. All rights reserved.
Déclaration de conflit d'intérêts
Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Références
PLoS Comput Biol. 2018 Jan 16;14(1):e1005896
pubmed: 29337990
JMLR Workshop Conf Proc. 2016 Aug;56:301-318
pubmed: 28286600
Adv Neural Inf Process Syst. 2015;28:3599-3607
pubmed: 27019571
J Biomed Inform. 2018 Oct;86:33-48
pubmed: 30138699
Am J Psychiatry. 2017 Feb 1;174(2):154-162
pubmed: 27609239
Health Aff (Millwood). 2018 May;37(5):694-701
pubmed: 29733705
PLoS One. 2016 May 02;11(5):e0154515
pubmed: 27135409
Pac Symp Biocomput. 2022;27:266-277
pubmed: 34890155
Health Serv Outcomes Res Methodol. 2017 Jun;17(2):101-112
pubmed: 28943779
J Biomed Inform. 2021 May;117:103698
pubmed: 33617985
Transl Psychiatry. 2020 Nov 26;10(1):413
pubmed: 33243979
J Biomed Inform. 2018 Oct;86:149-159
pubmed: 30172760
Am J Respir Crit Care Med. 2017 Oct 1;196(7):856-863
pubmed: 28345952
AMIA Jt Summits Transl Sci Proc. 2016 Jul 22;2016:428-37
pubmed: 27595047
Stat Med. 2017 Sep 30;36(22):3583-3595
pubmed: 28660629
IEEE J Biomed Health Inform. 2018 Sep;22(5):1589-1604
pubmed: 29989977
J Biomed Inform. 2018 Aug;84:11-16
pubmed: 29908902
Sci Rep. 2021 Oct 19;11(1):20685
pubmed: 34667200
JMIR Med Inform. 2020 Oct 21;8(10):e21798
pubmed: 33084589
Comput Biol Med. 2017 Oct 1;89:248-255
pubmed: 28843829
J Biomed Inform. 2015 Dec;58:156-165
pubmed: 26464024
Clin Microbiol Infect. 2020 Oct;26(10):1291-1299
pubmed: 32061798
J Biomed Inform. 2012 Dec;45(6):1108-19
pubmed: 22846170
Med Care. 2010 Jun;48(6 Suppl):S106-13
pubmed: 20473190
IEEE Trans Biomed Eng. 2018 Jan;65(1):207-218
pubmed: 28463183
Semin Radiat Oncol. 2019 Oct;29(4):354-361
pubmed: 31472738
J Biomed Inform. 2018 Feb;78:33-42
pubmed: 29196114
Nature. 2015 May 28;521(7553):452-9
pubmed: 26017444
Annu Rev Stat Appl. 2019 Mar;6:263-286
pubmed: 31073534