Prediction of in-hospital adverse clinical outcomes in patients with pulmonary thromboembolism, machine learning based models.
gradient boosting machine
logistic models
machine learing
outcome analysis
pulmonary embolism
risk factors
Journal
Frontiers in cardiovascular medicine
ISSN: 2297-055X
Titre abrégé: Front Cardiovasc Med
Pays: Switzerland
ID NLM: 101653388
Informations de publication
Date de publication:
2023
2023
Historique:
received:
02
11
2022
accepted:
27
02
2023
medline:
1
4
2023
entrez:
31
3
2023
pubmed:
1
4
2023
Statut:
epublish
Résumé
Pulmonary thromboembolism (PE) is the third leading cause of cardiovascular events. The conventional modeling methods and severity risk scores lack multiple laboratories, paraclinical and imaging data. Data science and machine learning (ML) based prediction models may help better predict outcomes. In this retrospective registry-based design, all consecutive hospitalized patients diagnosed with pulmonary thromboembolism (based on pulmonary CT angiography) from 2011 to 2019 were recruited. ML based algorithms [Gradient Boosting (GB) and Deep Learning (DL)] were applied and compared with logistic regression (LR) to predict hemodynamic instability and/or all-cause mortality. A total number of 1,017 patients were finally enrolled in the study, including 465 women and 552 men. Overall incidence of study main endpoint was 9.6%, (7.2% in men and 12.4% in women; ML-based models have notable prediction ability in PE patients. These algorithms may help physicians to detect high-risk patients earlier and take appropriate preventive measures.
Sections du résumé
Background
UNASSIGNED
Pulmonary thromboembolism (PE) is the third leading cause of cardiovascular events. The conventional modeling methods and severity risk scores lack multiple laboratories, paraclinical and imaging data. Data science and machine learning (ML) based prediction models may help better predict outcomes.
Materials and methods
UNASSIGNED
In this retrospective registry-based design, all consecutive hospitalized patients diagnosed with pulmonary thromboembolism (based on pulmonary CT angiography) from 2011 to 2019 were recruited. ML based algorithms [Gradient Boosting (GB) and Deep Learning (DL)] were applied and compared with logistic regression (LR) to predict hemodynamic instability and/or all-cause mortality.
Results
UNASSIGNED
A total number of 1,017 patients were finally enrolled in the study, including 465 women and 552 men. Overall incidence of study main endpoint was 9.6%, (7.2% in men and 12.4% in women;
Conclusion
UNASSIGNED
ML-based models have notable prediction ability in PE patients. These algorithms may help physicians to detect high-risk patients earlier and take appropriate preventive measures.
Identifiants
pubmed: 36998977
doi: 10.3389/fcvm.2023.1087702
pmc: PMC10043172
doi:
Types de publication
Journal Article
Langues
eng
Pagination
1087702Informations de copyright
© 2023 Jenab, Hosseini, Esmaeili, Tofighi, Ariannejad and Sotoudeh.
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 Thromb Haemost. 2018 Dec;16(12):2397-2402
pubmed: 30251461
J Am Coll Cardiol. 2001 Nov 15;38(6):1654-61
pubmed: 11704377
Rev Port Pneumol. 2014 Jul-Aug;20(4):194-9
pubmed: 24462346
Thromb Res. 2014 Nov;134(5):991-6
pubmed: 25228399
Thromb Haemost. 2000 Oct;84(4):548-52
pubmed: 11057848
Thromb Haemost. 2009 Jul;102(1):153-8
pubmed: 19572080
Crit Care Explor. 2021 Jun 11;3(6):e0437
pubmed: 34136820
Am J Med. 2018 Dec;131(12):1506-1514.e0
pubmed: 30102908
BMC Pulm Med. 2013 Dec 10;13:74
pubmed: 24325351
Lancet. 2012 May 12;379(9828):1835-46
pubmed: 22494827
Am J Emerg Med. 2017 Jan;35(1):146-149
pubmed: 27836322
J Int Med Res. 2021 Apr;49(4):3000605211004769
pubmed: 33823631
Libyan J Med. 2022 Dec;17(1):2044597
pubmed: 35227164
Arch Intern Med. 2010 Aug 9;170(15):1383-9
pubmed: 20696966
Eur Heart J. 2008 Jun;29(12):1569-77
pubmed: 18495689
PLoS Med. 2009 Sep;6(9):e1000145
pubmed: 19901973
Int J Cardiovasc Imaging. 2012 Apr;28(4):979-87
pubmed: 21717126
Lancet. 2017 Dec 17;388(10063):3060-3073
pubmed: 27375038
Medicina (B Aires). 2020;80(5):462-472
pubmed: 33048790
Ann Transl Med. 2019 Apr;7(7):152
pubmed: 31157273
J Intern Med. 2007 Jan;261(1):74-81
pubmed: 17222170
Semin Hematol. 2007 Apr;44(2):62-9
pubmed: 17433897
J Mol Cell Cardiol. 2006 Aug;41(2):296-307
pubmed: 16814320
Chest. 2011 Oct;140(4):1008-1015
pubmed: 21474573
Respir Res. 2012 Dec 04;13:111
pubmed: 23210843
Am J Hematol. 2013 Aug;88(8):677-81
pubmed: 23674436
Eur Heart J Acute Cardiovasc Care. 2021 May 11;10(3):250-257
pubmed: 33620435
Respirology. 2018 Oct;23(10):935-941
pubmed: 29693295
Arch Intern Med. 1997 Dec 8-22;157(22):2593-8
pubmed: 9531228
Eur Heart J. 2006 Feb;27(4):476-81
pubmed: 16207738
Acad Emerg Med. 2015 Oct;22(10):1127-37
pubmed: 26394330
Am J Respir Crit Care Med. 2005 Oct 15;172(8):1041-6
pubmed: 16020800
Clin Biochem. 2011 Sep;44(13):1110-1115
pubmed: 21723271
N Engl J Med. 1992 May 7;326(19):1240-5
pubmed: 1560799
Wien Klin Wochenschr. 2002 Sep 30;114(17-18):766-72
pubmed: 12416281
Chest. 2008 Feb;133(2):358-62
pubmed: 17951624
J Thromb Haemost. 2013 Dec;11(12):2092-102
pubmed: 24134450