Repository logo
 
Publication

Predicting ICU delirium in critically Ill COVID-19 patients using demographic, clinical, and laboratory admission data: a machine learning approach

2025-06Journal article Open access
Simple item page
datacite.subject.sdg03:Saúde de Qualidade
dc.contributor.authorViegas, Ana
dc.contributor.authorVon Rekowski, Cristiana P.
dc.contributor.authorAraújo, Rúben
dc.contributor.authorViana-Baptista, Miguel
dc.contributor.authorMacedo, Maria Paula
dc.contributor.authorBento, Luís
dc.date.accessioned2025-12-18T08:59:55Z
dc.date.available2025-12-18T08:59:55Z
dc.date.issued2025-06
dc.descriptionThis research was funded by the project grant DSAIPA/DS/0117/2020, supported by FCT—Fundação para a Ciência e Tecnologia, I.P. Cristiana P. Von Rekowski and Rúben Araújo acknowledge the support received from FCT through the PhD grants 2023.01951.BD (DOI: https://doi.org/10.54499/2023.01951.BD) and 2021.05553.BD (DOI: https://doi.org/10.54499/2021.05553.BD), respectively.
dc.description.abstractDelirium is a common and underrecognized complication among critically ill patients, associated with prolonged ICU stays, cognitive dysfunction, and increased mortality. Its multifactorial causes and fluctuating course hinder early prediction, limiting timely management. Predictive models based on data available at ICU admission may help to identify high-risk patients and guide early interventions. This study evaluated machine learning models used to predict delirium in critically ill patients with SARS-CoV-2 infections using a prospective cohort of 426 patients. The dataset included demographic characteristics, clinical data (e.g., comorbidities, medication, reason for ICU admission, interventions), and routine lab test results. Five models-Logistic Regression, Support Vector Machine, Decision Tree, Random Forest, and Naïve Bayes-were developed using 112 features. Feature selection relied on Information Gain, and model performance was assessed via 10-fold cross-validation. The Naïve Bayes model showed moderate predictive performance and high interpretability, achieving an AUC of 0.717, accuracy of 65.3%, sensitivity of 62.4%, specificity of 68.1%, and precision of 66.2%. Key predictors included invasive mechanical ventilation, deep sedation with benzodiazepines, SARS-CoV-2 as the reason for ICU admission, ECMO use, constipation, and male sex. These findings support the use of interpretable models for early delirium risk stratification using routinely available ICU data.eng
dc.identifier.citationViegas A, Von Rekowski CP, Araújo R, Viana-Baptista M, Macedo MP, Bento L. Predicting ICU delirium in critically Ill COVID-19 patients using demographic, clinical, and laboratory admission data: a machine learning approach. Life (Basel). 2025;15(7):1045.
dc.identifier.doi10.3390/life15071045
dc.identifier.issn2075-1729
dc.identifier.urihttp://hdl.handle.net/10400.21/22356
dc.language.isoeng
dc.peerreviewedyes
dc.publisherMDPI AG
dc.relation.hasversionhttps://www.mdpi.com/2075-1729/15/7/1045
dc.relation.ispartofLife
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/
dc.subjectCOVID-19
dc.subjectSARS-CoV-2
dc.subjectICU
dc.subjectDelirium
dc.subjectMachine learning
dc.subjectPredictive modeling
dc.subjectIntensive care unit
dc.titlePredicting ICU delirium in critically Ill COVID-19 patients using demographic, clinical, and laboratory admission data: a machine learning approacheng
dc.typejournal article
dcterms.referenceshttps://ww w.mdpi.com/article/10.3390/life15071045/s1
dspace.entity.typePublication
oaire.citation.issue7
oaire.citation.startPage1045
oaire.citation.titleLife
oaire.citation.volume15
oaire.versionhttp://purl.org/coar/version/c_970fb48d4fbd8a85

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Predicting ICU delirium in critically Ill COVID-19 patients using demographic, clinical, and laboratory admission data_a machine learning approach.pdf
Size:
609.27 KB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
4.03 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

ESTeSL - Artigos