Complete blood count and C-reactive protein to predict positive blood culture among neonates using machine learning algorithms
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Clinics |
Texto Completo: | https://www.revistas.usp.br/clinics/article/view/213656 |
Resumo: | Purpose: The authors aimed to develop a Machine-Learning (ML) algorithm that can predict positive blood culture in the neonatal intensive care unit, using complete blood count and C-reactive protein values. Methods: The study was based on patients’ electronic health records at a tertiary neonatal intensive care unit in São Paulo, Brazil. All blood cultures that had paired complete blood count and C-reactive protein measurements taken at the same time were included. To evaluate the machine learning model's performance, the authors used accuracy, Area Under the Receiver Operating Characteristics (AUROC), recall, precision, and F1-score. Results: The dataset included 1181 blood cultures with paired complete blood count plus c-reactive protein and 1911 blood cultures with paired complete blood count only. The f1-score ranged from 0.14 to 0.43, recall ranged from 0.08 to 0.59, precision ranged from 0.29 to 1.00, and accuracy ranged from 0.688 to 0.864. Conclusion: Complete blood count parameters and C-reactive protein levels cannot be used in ML models to predict bacteremia in newborns. |
id |
USP-19_967566dff2ff6b430a47ef3b7e81fc70 |
---|---|
oai_identifier_str |
oai:revistas.usp.br:article/213656 |
network_acronym_str |
USP-19 |
network_name_str |
Clinics |
repository_id_str |
|
spelling |
Complete blood count and C-reactive protein to predict positive blood culture among neonates using machine learning algorithmsCritical careNeonatologyArtificial intelligenceMachine learningSepsisPurpose: The authors aimed to develop a Machine-Learning (ML) algorithm that can predict positive blood culture in the neonatal intensive care unit, using complete blood count and C-reactive protein values. Methods: The study was based on patients’ electronic health records at a tertiary neonatal intensive care unit in São Paulo, Brazil. All blood cultures that had paired complete blood count and C-reactive protein measurements taken at the same time were included. To evaluate the machine learning model's performance, the authors used accuracy, Area Under the Receiver Operating Characteristics (AUROC), recall, precision, and F1-score. Results: The dataset included 1181 blood cultures with paired complete blood count plus c-reactive protein and 1911 blood cultures with paired complete blood count only. The f1-score ranged from 0.14 to 0.43, recall ranged from 0.08 to 0.59, precision ranged from 0.29 to 1.00, and accuracy ranged from 0.688 to 0.864. Conclusion: Complete blood count parameters and C-reactive protein levels cannot be used in ML models to predict bacteremia in newborns.Hospital das Clínicas, Faculdade de Medicina, Universidade de São Paulo2022-12-08info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://www.revistas.usp.br/clinics/article/view/21365610.1016/j.clinsp.2022.100148Clinics; Vol. 78 (2023); 100148Clinics; v. 78 (2023); 100148Clinics; Vol. 78 (2023); 1001481980-53221807-5932reponame:Clinicsinstname:Universidade de São Paulo (USP)instacron:USPenghttps://www.revistas.usp.br/clinics/article/view/213656/195758Copyright (c) 2023 Clinicsinfo:eu-repo/semantics/openAccessMatsushita, Felipe YuKrebs, Vera Lúcia JornadaCarvalho, Werther Brunow de2023-07-06T13:05:37Zoai:revistas.usp.br:article/213656Revistahttps://www.revistas.usp.br/clinicsPUBhttps://www.revistas.usp.br/clinics/oai||clinics@hc.fm.usp.br1980-53221807-5932opendoar:2023-07-06T13:05:37Clinics - Universidade de São Paulo (USP)false |
dc.title.none.fl_str_mv |
Complete blood count and C-reactive protein to predict positive blood culture among neonates using machine learning algorithms |
title |
Complete blood count and C-reactive protein to predict positive blood culture among neonates using machine learning algorithms |
spellingShingle |
Complete blood count and C-reactive protein to predict positive blood culture among neonates using machine learning algorithms Matsushita, Felipe Yu Critical care Neonatology Artificial intelligence Machine learning Sepsis |
title_short |
Complete blood count and C-reactive protein to predict positive blood culture among neonates using machine learning algorithms |
title_full |
Complete blood count and C-reactive protein to predict positive blood culture among neonates using machine learning algorithms |
title_fullStr |
Complete blood count and C-reactive protein to predict positive blood culture among neonates using machine learning algorithms |
title_full_unstemmed |
Complete blood count and C-reactive protein to predict positive blood culture among neonates using machine learning algorithms |
title_sort |
Complete blood count and C-reactive protein to predict positive blood culture among neonates using machine learning algorithms |
author |
Matsushita, Felipe Yu |
author_facet |
Matsushita, Felipe Yu Krebs, Vera Lúcia Jornada Carvalho, Werther Brunow de |
author_role |
author |
author2 |
Krebs, Vera Lúcia Jornada Carvalho, Werther Brunow de |
author2_role |
author author |
dc.contributor.author.fl_str_mv |
Matsushita, Felipe Yu Krebs, Vera Lúcia Jornada Carvalho, Werther Brunow de |
dc.subject.por.fl_str_mv |
Critical care Neonatology Artificial intelligence Machine learning Sepsis |
topic |
Critical care Neonatology Artificial intelligence Machine learning Sepsis |
description |
Purpose: The authors aimed to develop a Machine-Learning (ML) algorithm that can predict positive blood culture in the neonatal intensive care unit, using complete blood count and C-reactive protein values. Methods: The study was based on patients’ electronic health records at a tertiary neonatal intensive care unit in São Paulo, Brazil. All blood cultures that had paired complete blood count and C-reactive protein measurements taken at the same time were included. To evaluate the machine learning model's performance, the authors used accuracy, Area Under the Receiver Operating Characteristics (AUROC), recall, precision, and F1-score. Results: The dataset included 1181 blood cultures with paired complete blood count plus c-reactive protein and 1911 blood cultures with paired complete blood count only. The f1-score ranged from 0.14 to 0.43, recall ranged from 0.08 to 0.59, precision ranged from 0.29 to 1.00, and accuracy ranged from 0.688 to 0.864. Conclusion: Complete blood count parameters and C-reactive protein levels cannot be used in ML models to predict bacteremia in newborns. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-12-08 |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://www.revistas.usp.br/clinics/article/view/213656 10.1016/j.clinsp.2022.100148 |
url |
https://www.revistas.usp.br/clinics/article/view/213656 |
identifier_str_mv |
10.1016/j.clinsp.2022.100148 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
https://www.revistas.usp.br/clinics/article/view/213656/195758 |
dc.rights.driver.fl_str_mv |
Copyright (c) 2023 Clinics info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Copyright (c) 2023 Clinics |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Hospital das Clínicas, Faculdade de Medicina, Universidade de São Paulo |
publisher.none.fl_str_mv |
Hospital das Clínicas, Faculdade de Medicina, Universidade de São Paulo |
dc.source.none.fl_str_mv |
Clinics; Vol. 78 (2023); 100148 Clinics; v. 78 (2023); 100148 Clinics; Vol. 78 (2023); 100148 1980-5322 1807-5932 reponame:Clinics instname:Universidade de São Paulo (USP) instacron:USP |
instname_str |
Universidade de São Paulo (USP) |
instacron_str |
USP |
institution |
USP |
reponame_str |
Clinics |
collection |
Clinics |
repository.name.fl_str_mv |
Clinics - Universidade de São Paulo (USP) |
repository.mail.fl_str_mv |
||clinics@hc.fm.usp.br |
_version_ |
1800222767090499584 |