Using topological data analysis and pseudo time series to infer temporal phenotypes from electronic health records.
Electronic phenotyping
Longitudinal studies
Type 2 diabetes
Unsupervised machine learning
Journal
Artificial intelligence in medicine
ISSN: 1873-2860
Titre abrégé: Artif Intell Med
Pays: Netherlands
ID NLM: 8915031
Informations de publication
Date de publication:
08 2020
08 2020
Historique:
received:
06
01
2020
revised:
21
05
2020
accepted:
11
07
2020
entrez:
25
9
2020
pubmed:
26
9
2020
medline:
19
8
2021
Statut:
ppublish
Résumé
Temporal phenotyping enables clinicians to better understand observable characteristics of a disease as it progresses. Modelling disease progression that captures interactions between phenotypes is inherently challenging. Temporal models that capture change in disease over time can identify the key features that characterize disease subtypes that underpin these trajectories. These models will enable clinicians to identify early warning signs of progression in specific sub-types and therefore to make informed decisions tailored to individual patients. In this paper, we explore two approaches to building temporal phenotypes based on the topology of data: topological data analysis and pseudo time-series. Using type 2 diabetes data, we show that the topological data analysis approach is able to identify disease trajectories and that pseudo time-series can infer a state space model characterized by transitions between hidden states that represent distinct temporal phenotypes. Both approaches highlight lipid profiles as key factors in distinguishing the phenotypes.
Identifiants
pubmed: 32972659
pii: S0933-3657(19)31184-4
doi: 10.1016/j.artmed.2020.101930
pmc: PMC7536308
pii:
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
101930Subventions
Organisme : Medical Research Council
ID : MR/N00583X/1
Pays : United Kingdom
Informations de copyright
Copyright © 2020 The Authors. Published by Elsevier B.V. All rights reserved.
Références
Anal Chim Acta. 2016 Mar 3;910:1-11
pubmed: 26873463
J Am Med Inform Assoc. 2018 May 1;25(5):538-547
pubmed: 29409033
Artif Intell Med. 2017 Mar;77:23-30
pubmed: 28545609
J Am Med Inform Assoc. 2013 Jan 1;20(1):117-21
pubmed: 22955496
J Diabetes Sci Technol. 2018 Mar;12(2):251-259
pubmed: 29493360
J Diabetes Sci Technol. 2018 Mar;12(2):295-302
pubmed: 28494618
J Biomed Inform. 2013 Apr;46(2):266-74
pubmed: 23200810
JAMA. 2018 Dec 4;320(21):2199-2200
pubmed: 30398550
ACM Trans Intell Syst Technol. 2013 Sep;4(4):
pubmed: 25309815
J Biomed Inform. 2017 Feb;66:136-147
pubmed: 28057564
Nat Commun. 2018 Jun 22;9(1):2442
pubmed: 29934517
Bioinformatics. 2003 May 1;19(7):842-50
pubmed: 12724294
KDD. 2012;2012:280-288
pubmed: 25937993
Sci Rep. 2013;3:1236
pubmed: 23393618
IEEE/ACM Trans Comput Biol Bioinform. 2008 Apr-Jun;5(2):172-82
pubmed: 18451427
Proc Natl Acad Sci U S A. 2011 Apr 26;108(17):7265-70
pubmed: 21482760
IEEE Trans Inf Technol Biomed. 2010 Jan;14(1):79-85
pubmed: 19527963
BMC Bioinformatics. 2020 Jul 29;21(1):336
pubmed: 32727348
Cell Syst. 2017 Nov 22;5(5):527-531.e3
pubmed: 29102608
Diab Vasc Dis Res. 2018 Sep;15(5):424-432
pubmed: 29911415
Sci Transl Med. 2015 Oct 28;7(311):311ra174
pubmed: 26511511