Man vs. machine : predicting hospital bed demand
Autor(a) principal: | |
---|---|
Data de Publicação: | 2020 |
Outros Autores: | , , , , , , , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UFRGS |
Texto Completo: | http://hdl.handle.net/10183/220309 |
Resumo: | Background: The recent literature reports promising results from using intelligent systems to support decision making in healthcare operations. Using these systems may lead to improved diagnostic and treatment protocols and to predict hospital bed demand. Predicting hospital bed demand in emergency department (ED) attendances could help resource allocation and reduce pressure on busy hospitals. However, there is still limited knowledge on whether intelligent systems can operate as fully autonomous, user-independent systems. Objective: Compare the performance of a computer-based algorithm and humans in predicting hospital bed demand (admissions and discharges) based on the initial SOAP (Subjective, Objective, Assessment, Plan) records of the ED. Methods: This was a retrospective cohort study that compared the performance of humans and machines in predicting hospital bed demand from an ED. It considered electronic medical records (EMR) of 9030 patients (230 used as a testing set, and hence evaluated both by humans and by an algorithm, and 8800 used as a training set exclusively by the algorithm) who visited the ED of a tertiary care and teaching public hospital located in Porto Alegre, Brazil between January and December 2014. The machine role was played by Support Vector Machine Classifier and the human prediction was performed by four ED physicians. Predictions were compared in terms of sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUROC). Results: All graders achieved similar accuracies. The accuracy by AUROC for the testing set was 0.82 [95% confidence interval (CI) of 0.77–0.87], 0.80 (95% CI: 0.75–0.85), 0.76 (95% CI: 0.71–0.81) for novice physicians, machine, experienced physicians, respectively. Processing time per test EMR was 0.00812±0.0009 seconds. In contrast, novice physicians took on average 156.80 seconds per test EMR, while experienced physicians took on average 56.40 seconds per test EMR. Conclusions: Our data indicated that the system could predict patient admission or discharge states with 80% accuracy, which was similar the performance of novice and experienced physicians. These results suggested that the algorithm could operate as an autonomous and independent system to complete this task. |
id |
UFRGS-2_bfc2a2b807569b61710bc9305ddff47d |
---|---|
oai_identifier_str |
oai:www.lume.ufrgs.br:10183/220309 |
network_acronym_str |
UFRGS-2 |
network_name_str |
Repositório Institucional da UFRGS |
repository_id_str |
|
spelling |
Lucini, Filipe RissieriReis, Mateus Augusto dosSilveira, Giovani Jose Caetano daFogliatto, Flavio SansonAnzanello, Michel JoséAndrioli, Giordanna GuerraNicolaidis, RafaelBeltrame, Rafael Coimbra FerreiraNeyeloff, Jeruza LavanholiSchaan, Beatriz D'Agord2021-04-28T04:31:37Z20201932-6203http://hdl.handle.net/10183/220309001123833Background: The recent literature reports promising results from using intelligent systems to support decision making in healthcare operations. Using these systems may lead to improved diagnostic and treatment protocols and to predict hospital bed demand. Predicting hospital bed demand in emergency department (ED) attendances could help resource allocation and reduce pressure on busy hospitals. However, there is still limited knowledge on whether intelligent systems can operate as fully autonomous, user-independent systems. Objective: Compare the performance of a computer-based algorithm and humans in predicting hospital bed demand (admissions and discharges) based on the initial SOAP (Subjective, Objective, Assessment, Plan) records of the ED. Methods: This was a retrospective cohort study that compared the performance of humans and machines in predicting hospital bed demand from an ED. It considered electronic medical records (EMR) of 9030 patients (230 used as a testing set, and hence evaluated both by humans and by an algorithm, and 8800 used as a training set exclusively by the algorithm) who visited the ED of a tertiary care and teaching public hospital located in Porto Alegre, Brazil between January and December 2014. The machine role was played by Support Vector Machine Classifier and the human prediction was performed by four ED physicians. Predictions were compared in terms of sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUROC). Results: All graders achieved similar accuracies. The accuracy by AUROC for the testing set was 0.82 [95% confidence interval (CI) of 0.77–0.87], 0.80 (95% CI: 0.75–0.85), 0.76 (95% CI: 0.71–0.81) for novice physicians, machine, experienced physicians, respectively. Processing time per test EMR was 0.00812±0.0009 seconds. In contrast, novice physicians took on average 156.80 seconds per test EMR, while experienced physicians took on average 56.40 seconds per test EMR. Conclusions: Our data indicated that the system could predict patient admission or discharge states with 80% accuracy, which was similar the performance of novice and experienced physicians. These results suggested that the algorithm could operate as an autonomous and independent system to complete this task.application/pdfengPLoS One. San Francisco. vol. 15, no. 8 (Aug. 2020), e0237937, 11 p.Ocupação de leitosServiço hospitalar de emergênciaPrevisõesPhysiciansCritical care and emergency medicineInpatientsMachine learning algorithmsAlgorithmsElectronic medical recordsSupport vector machinesHospitalsMan vs. machine : predicting hospital bed demandEstrangeiroinfo:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/openAccessreponame:Repositório Institucional da UFRGSinstname:Universidade Federal do Rio Grande do Sul (UFRGS)instacron:UFRGSTEXT001123833.pdf.txt001123833.pdf.txtExtracted Texttext/plain37829http://www.lume.ufrgs.br/bitstream/10183/220309/2/001123833.pdf.txte9bb783d73dc3871a2ecd25a7a32f5e5MD52ORIGINAL001123833.pdfTexto completo (inglês)application/pdf917885http://www.lume.ufrgs.br/bitstream/10183/220309/1/001123833.pdf0467dc5928b4e70928d4f2effdda6472MD5110183/2203092023-09-24 03:39:42.037986oai:www.lume.ufrgs.br:10183/220309Repositório InstitucionalPUBhttps://lume.ufrgs.br/oai/requestlume@ufrgs.bropendoar:2023-09-24T06:39:42Repositório Institucional da UFRGS - Universidade Federal do Rio Grande do Sul (UFRGS)false |
dc.title.pt_BR.fl_str_mv |
Man vs. machine : predicting hospital bed demand |
title |
Man vs. machine : predicting hospital bed demand |
spellingShingle |
Man vs. machine : predicting hospital bed demand Lucini, Filipe Rissieri Ocupação de leitos Serviço hospitalar de emergência Previsões Physicians Critical care and emergency medicine Inpatients Machine learning algorithms Algorithms Electronic medical records Support vector machines Hospitals |
title_short |
Man vs. machine : predicting hospital bed demand |
title_full |
Man vs. machine : predicting hospital bed demand |
title_fullStr |
Man vs. machine : predicting hospital bed demand |
title_full_unstemmed |
Man vs. machine : predicting hospital bed demand |
title_sort |
Man vs. machine : predicting hospital bed demand |
author |
Lucini, Filipe Rissieri |
author_facet |
Lucini, Filipe Rissieri Reis, Mateus Augusto dos Silveira, Giovani Jose Caetano da Fogliatto, Flavio Sanson Anzanello, Michel José Andrioli, Giordanna Guerra Nicolaidis, Rafael Beltrame, Rafael Coimbra Ferreira Neyeloff, Jeruza Lavanholi Schaan, Beatriz D'Agord |
author_role |
author |
author2 |
Reis, Mateus Augusto dos Silveira, Giovani Jose Caetano da Fogliatto, Flavio Sanson Anzanello, Michel José Andrioli, Giordanna Guerra Nicolaidis, Rafael Beltrame, Rafael Coimbra Ferreira Neyeloff, Jeruza Lavanholi Schaan, Beatriz D'Agord |
author2_role |
author author author author author author author author author |
dc.contributor.author.fl_str_mv |
Lucini, Filipe Rissieri Reis, Mateus Augusto dos Silveira, Giovani Jose Caetano da Fogliatto, Flavio Sanson Anzanello, Michel José Andrioli, Giordanna Guerra Nicolaidis, Rafael Beltrame, Rafael Coimbra Ferreira Neyeloff, Jeruza Lavanholi Schaan, Beatriz D'Agord |
dc.subject.por.fl_str_mv |
Ocupação de leitos Serviço hospitalar de emergência Previsões |
topic |
Ocupação de leitos Serviço hospitalar de emergência Previsões Physicians Critical care and emergency medicine Inpatients Machine learning algorithms Algorithms Electronic medical records Support vector machines Hospitals |
dc.subject.eng.fl_str_mv |
Physicians Critical care and emergency medicine Inpatients Machine learning algorithms Algorithms Electronic medical records Support vector machines Hospitals |
description |
Background: The recent literature reports promising results from using intelligent systems to support decision making in healthcare operations. Using these systems may lead to improved diagnostic and treatment protocols and to predict hospital bed demand. Predicting hospital bed demand in emergency department (ED) attendances could help resource allocation and reduce pressure on busy hospitals. However, there is still limited knowledge on whether intelligent systems can operate as fully autonomous, user-independent systems. Objective: Compare the performance of a computer-based algorithm and humans in predicting hospital bed demand (admissions and discharges) based on the initial SOAP (Subjective, Objective, Assessment, Plan) records of the ED. Methods: This was a retrospective cohort study that compared the performance of humans and machines in predicting hospital bed demand from an ED. It considered electronic medical records (EMR) of 9030 patients (230 used as a testing set, and hence evaluated both by humans and by an algorithm, and 8800 used as a training set exclusively by the algorithm) who visited the ED of a tertiary care and teaching public hospital located in Porto Alegre, Brazil between January and December 2014. The machine role was played by Support Vector Machine Classifier and the human prediction was performed by four ED physicians. Predictions were compared in terms of sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUROC). Results: All graders achieved similar accuracies. The accuracy by AUROC for the testing set was 0.82 [95% confidence interval (CI) of 0.77–0.87], 0.80 (95% CI: 0.75–0.85), 0.76 (95% CI: 0.71–0.81) for novice physicians, machine, experienced physicians, respectively. Processing time per test EMR was 0.00812±0.0009 seconds. In contrast, novice physicians took on average 156.80 seconds per test EMR, while experienced physicians took on average 56.40 seconds per test EMR. Conclusions: Our data indicated that the system could predict patient admission or discharge states with 80% accuracy, which was similar the performance of novice and experienced physicians. These results suggested that the algorithm could operate as an autonomous and independent system to complete this task. |
publishDate |
2020 |
dc.date.issued.fl_str_mv |
2020 |
dc.date.accessioned.fl_str_mv |
2021-04-28T04:31:37Z |
dc.type.driver.fl_str_mv |
Estrangeiro info:eu-repo/semantics/article |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/10183/220309 |
dc.identifier.issn.pt_BR.fl_str_mv |
1932-6203 |
dc.identifier.nrb.pt_BR.fl_str_mv |
001123833 |
identifier_str_mv |
1932-6203 001123833 |
url |
http://hdl.handle.net/10183/220309 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.ispartof.pt_BR.fl_str_mv |
PLoS One. San Francisco. vol. 15, no. 8 (Aug. 2020), e0237937, 11 p. |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional da UFRGS instname:Universidade Federal do Rio Grande do Sul (UFRGS) instacron:UFRGS |
instname_str |
Universidade Federal do Rio Grande do Sul (UFRGS) |
instacron_str |
UFRGS |
institution |
UFRGS |
reponame_str |
Repositório Institucional da UFRGS |
collection |
Repositório Institucional da UFRGS |
bitstream.url.fl_str_mv |
http://www.lume.ufrgs.br/bitstream/10183/220309/2/001123833.pdf.txt http://www.lume.ufrgs.br/bitstream/10183/220309/1/001123833.pdf |
bitstream.checksum.fl_str_mv |
e9bb783d73dc3871a2ecd25a7a32f5e5 0467dc5928b4e70928d4f2effdda6472 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 |
repository.name.fl_str_mv |
Repositório Institucional da UFRGS - Universidade Federal do Rio Grande do Sul (UFRGS) |
repository.mail.fl_str_mv |
lume@ufrgs.br |
_version_ |
1817725094891356160 |