Prediction of Atrial Fibrillation in Hospitalized Elderly Patients With Coronary Heart Disease and Type 2 Diabetes Mellitus Using Machine Learning: A Multicenter Retrospective Study.
atrial fibrillation
coronary heart disease
machine learning
prediction models
type 2 diabetes mellitus
Journal
Frontiers in public health
ISSN: 2296-2565
Titre abrégé: Front Public Health
Pays: Switzerland
ID NLM: 101616579
Informations de publication
Date de publication:
2022
2022
Historique:
received:
23
12
2021
accepted:
09
02
2022
entrez:
21
3
2022
pubmed:
22
3
2022
medline:
29
4
2022
Statut:
epublish
Résumé
The objective of this study was to use machine learning algorithms to construct predictive models for atrial fibrillation (AF) in elderly patients with coronary heart disease (CHD) and type 2 diabetes mellitus (T2DM). The diagnosis and treatment data of elderly patients with CHD and T2DM, who were treated in four tertiary hospitals in Chongqing, China from 2015 to 2021, were collected. Five machine learning algorithms: logistic regression, logistic regression+least absolute shrinkage and selection operator, classified regression tree (CART), random forest (RF) and extreme gradient lifting (XGBoost) were used to construct the prediction models. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy were used as the comparison measures between different models. A total of 3,858 elderly patients with CHD and T2DM were included. In the internal validation cohort, XGBoost had the highest AUC (0.743) and sensitivity (0.833), and RF had the highest specificity (0.753) and accuracy (0.735). In the external verification, RF had the highest AUC (0.726) and sensitivity (0.686), and CART had the highest specificity (0.925) and accuracy (0.841). Total bilirubin, triglycerides and uric acid were the three most important predictors of AF. The risk prediction models of AF in elderly patients with CHD and T2DM based on machine learning algorithms had high diagnostic value. The prediction models constructed by RF and XGBoost were more effective. The results of this study can provide reference for the clinical prevention and treatment of AF.
Sections du résumé
Background
The objective of this study was to use machine learning algorithms to construct predictive models for atrial fibrillation (AF) in elderly patients with coronary heart disease (CHD) and type 2 diabetes mellitus (T2DM).
Methods
The diagnosis and treatment data of elderly patients with CHD and T2DM, who were treated in four tertiary hospitals in Chongqing, China from 2015 to 2021, were collected. Five machine learning algorithms: logistic regression, logistic regression+least absolute shrinkage and selection operator, classified regression tree (CART), random forest (RF) and extreme gradient lifting (XGBoost) were used to construct the prediction models. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy were used as the comparison measures between different models.
Results
A total of 3,858 elderly patients with CHD and T2DM were included. In the internal validation cohort, XGBoost had the highest AUC (0.743) and sensitivity (0.833), and RF had the highest specificity (0.753) and accuracy (0.735). In the external verification, RF had the highest AUC (0.726) and sensitivity (0.686), and CART had the highest specificity (0.925) and accuracy (0.841). Total bilirubin, triglycerides and uric acid were the three most important predictors of AF.
Conclusion
The risk prediction models of AF in elderly patients with CHD and T2DM based on machine learning algorithms had high diagnostic value. The prediction models constructed by RF and XGBoost were more effective. The results of this study can provide reference for the clinical prevention and treatment of AF.
Identifiants
pubmed: 35309227
doi: 10.3389/fpubh.2022.842104
pmc: PMC8931193
doi:
Types de publication
Journal Article
Multicenter Study
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
842104Informations de copyright
Copyright © 2022 Xu, Peng, Tan, Zhao, Yang and Tian.
Déclaration de conflit d'intérêts
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Références
J Pers Med. 2020 Aug 09;10(3):
pubmed: 32784873
J Biomed Inform. 2002 Oct-Dec;35(5-6):352-9
pubmed: 12968784
Int J Cardiol. 2016 Nov 1;222:1079-1083
pubmed: 27514627
PeerJ. 2021 Feb 17;9:e10903
pubmed: 33643714
Eur J Cardiovasc Prev Rehabil. 2007 Feb;14(1):79-84
pubmed: 17301631
Metabolites. 2020 May 14;10(5):
pubmed: 32423050
Circulation. 2006 Jul 11;114(2):119-25
pubmed: 16818816
Diabetes Metab Syndr Obes. 2020 Dec 18;13:5025-5036
pubmed: 33376372
Front Cardiovasc Med. 2021 Feb 02;8:614204
pubmed: 33634169
Neurol Res. 2015 Aug;37(8):727-31
pubmed: 25891436
Int J Cardiol. 2017 Mar 15;231:137-142
pubmed: 27871785
Liver Int. 2019 Dec;39(12):2341-2349
pubmed: 31436903
Intern Med. 2011;50(8):799-803
pubmed: 21498925
J Clin Lipidol. 2018 Mar - Apr;12(2):356-366
pubmed: 29310989
Am J Physiol Heart Circ Physiol. 2021 Jan 1;320(1):H1-H12
pubmed: 33185113
Int J Comput Assist Radiol Surg. 2021 Dec;16(12):2235-2249
pubmed: 34677748
Front Aging Neurosci. 2017 Oct 06;9:329
pubmed: 29056906
J Am Coll Cardiol. 2006 Aug 15;48(4):854-906
pubmed: 16904574
Clin Cardiol. 2012 May;35(5):301-6
pubmed: 22262261
Asian J Androl. 2017 Sep-Oct;19(5):586-590
pubmed: 27586028
Diabetes Metab Res Rev. 2018 Feb;34(2):
pubmed: 29124871
BMJ Open. 2020 Nov 27;10(11):e039236
pubmed: 33247009
Circ J. 2018 Oct 25;82(11):2728-2735
pubmed: 30232315
Comput Biol Med. 2021 Oct;137:104813
pubmed: 34481185
Annu Rev Med. 2022 Jan 27;73:355-362
pubmed: 34788544
Clin Sci (Lond). 2015 Jul;129(1):1-25
pubmed: 25881719
J Am Coll Cardiol. 2021 Jun 22;77(24):3031-3041
pubmed: 34140107
Int J Cardiol. 2014 Jun 15;174(2):293-8
pubmed: 24794549
Eur J Epidemiol. 2014 Mar;29(3):181-90
pubmed: 24389686
JAMA. 2013 Nov 20;310(19):2050-60
pubmed: 24240932
Hepatology. 2016 Jul;64(1):200-8
pubmed: 26690389
Trends Cardiovasc Med. 2017 Aug;27(6):428-432
pubmed: 28438398