Machine learning-based prediction of fainting during blood donations using donor properties and weather data as features.
Blood donation
Donor safety
Fainting
Journal
BMC medical informatics and decision making
ISSN: 1472-6947
Titre abrégé: BMC Med Inform Decis Mak
Pays: England
ID NLM: 101088682
Informations de publication
Date de publication:
20 08 2022
20 08 2022
Historique:
received:
05
02
2022
accepted:
08
08
2022
entrez:
20
8
2022
pubmed:
21
8
2022
medline:
24
8
2022
Statut:
epublish
Résumé
Fainting is a well-known side effect of blood donation. Such adverse experiences can diminish the return rate for further blood donations. Identifying factors associated with fainting could help prevent adverse incidents during blood donation. Data of 85,040 blood donations from whole blood and apheresis donors within four consecutive years were included in this retrospective study. Seven different machine learning models (random forests, artificial neural networks, XGradient Boosting, AdaBoost, logistic regression, K nearest neighbors, and support vector machines) for predicting fainting during blood donation were established. The used features derived from the data obtained from the questionnaire every donor has to fill in before the donation and weather data of the day of the donation. One thousand seven hundred fifteen fainting reactions were observed in 228 846 blood donations from 88,003 donors over a study period of 48 months. Similar values for all machine learning algorithms investigated for NPV, PPV, AUC, and F1-score were obtained. In general, NPV was above 0.996, whereas PPV was below 0.03. AUC and F1-score were close to 0.9 for all models. Essential features predicting fainting during blood donation were systolic and diastolic blood pressure and ambient temperature, humidity, and barometric pressure. Machine-learning algorithms can establish prediction models of fainting in blood donors. These new tools can reduce adverse reactions during blood donation and improve donor safety and minimize negative associations relating to blood donation.
Sections du résumé
BACKGROUND AND OBJECTIVES
Fainting is a well-known side effect of blood donation. Such adverse experiences can diminish the return rate for further blood donations. Identifying factors associated with fainting could help prevent adverse incidents during blood donation.
MATERIALS AND METHODS
Data of 85,040 blood donations from whole blood and apheresis donors within four consecutive years were included in this retrospective study. Seven different machine learning models (random forests, artificial neural networks, XGradient Boosting, AdaBoost, logistic regression, K nearest neighbors, and support vector machines) for predicting fainting during blood donation were established. The used features derived from the data obtained from the questionnaire every donor has to fill in before the donation and weather data of the day of the donation.
RESULTS
One thousand seven hundred fifteen fainting reactions were observed in 228 846 blood donations from 88,003 donors over a study period of 48 months. Similar values for all machine learning algorithms investigated for NPV, PPV, AUC, and F1-score were obtained. In general, NPV was above 0.996, whereas PPV was below 0.03. AUC and F1-score were close to 0.9 for all models. Essential features predicting fainting during blood donation were systolic and diastolic blood pressure and ambient temperature, humidity, and barometric pressure.
CONCLUSION
Machine-learning algorithms can establish prediction models of fainting in blood donors. These new tools can reduce adverse reactions during blood donation and improve donor safety and minimize negative associations relating to blood donation.
Identifiants
pubmed: 35987636
doi: 10.1186/s12911-022-01971-x
pii: 10.1186/s12911-022-01971-x
pmc: PMC9392313
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
222Informations de copyright
© 2022. The Author(s).
Références
Circ Res. 1971 Oct;29(4):424-31
pubmed: 5110922
Transfusion. 2012 Sep;52(9):1871-9
pubmed: 22321030
Transfusion. 2022 Apr;62(4):838-847
pubmed: 35191034
Transfusion. 1992 Jan;32(1):23-6
pubmed: 1731431
Vox Sang. 2003 Feb;84(2):120-8
pubmed: 12609018
Am J Med. 2014 Aug;127(8):787-90
pubmed: 24613712
Transfusion. 2006 Feb;46(2):284-8
pubmed: 16441608
Blood Transfus. 2015 Jul;13(3):354-62
pubmed: 25761310
Clin Sci (Lond). 1990 Jul;79(1):73-9
pubmed: 2167794
Transfusion. 1998 Apr;38(4):359-67
pubmed: 9595018
Transfusion. 2006 Aug;46(8):1374-9
pubmed: 16934074
Atherosclerosis. 2002 Apr;161(2):395-402
pubmed: 11888523
Blood Transfus. 2014 Jan;12 Suppl 1:s28-36
pubmed: 23867173
Transfusion. 2002 Dec;42(12):1557-60
pubmed: 12473134
Vox Sang. 2016 Oct;111(3):257-265
pubmed: 27232998
Transfus Apher Sci. 2015 Aug;53(1):76-81
pubmed: 25824704
Transfus Apher Sci. 2012 Dec;47(3):319-25
pubmed: 22647682
Vox Sang. 2011 Nov;101(4):303-12
pubmed: 21535440
Transfusion. 2007 Nov;47(11):1999-2010
pubmed: 17958528
Transfusion. 2008 Sep;48(9):1809-19
pubmed: 18631167
PLoS One. 2017 Jul 21;12(7):e0179831
pubmed: 28732090
Vox Sang. 2016 Feb;110(2):107-15
pubmed: 26275052
Stat Anal Data Min. 2017 Dec;10(6):363-377
pubmed: 29403567
Transfusion. 2008 Sep;48(9):1799-808
pubmed: 18482188
Heart. 1997 Aug;78(2):188-93
pubmed: 9326996
Transfus Clin Biol. 2015 Oct-Dec;22(5-6):312-7
pubmed: 26476508
PLoS One. 2011;6(7):e22719
pubmed: 21818372
Neural Comput. 2006 Jun;18(6):1472-510
pubmed: 16764511
Medicine (Baltimore). 2019 Nov;98(44):e17758
pubmed: 31689834
Transfus Med. 2005 Oct;15(5):389-94
pubmed: 16202053
Transfusion. 2010 Jun;50(6):1257-64
pubmed: 20113455
Transfusion. 2019 May;59(5):1723-1733
pubmed: 30779173
Transfusion. 1999 Mar;39(3):316-20
pubmed: 10204597
Europace. 2019 Mar 1;21(3):511-521
pubmed: 30137300
Transfus Apher Sci. 2005 Oct;33(2):99-106
pubmed: 16103014