EUVIMEDEUVIMED entdecken
EvaluationsstudieOpen Access · CC BY 4.0

Evaluation of machine learning pipeline for blood culture outcome prediction on prospectively collected emergency department data

Benjamin R McFadden, Mark Reynolds, Antonio Celenza, Timothy J J Inglis

Journal of Medical Microbiology · 29.07.2026 · 75 (7)

DOI: 10.1099/jmm.0.002191 ↗

Kurz gefasst

Evaluationsstudie in der Zeitschrift Journal of Medical Microbiology, erschienen am 29.07.2026. Verfasst von Benjamin R McFadden und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Physics, Mathematics and Computing, University of Western Australia, Perth 6009, Australia. Hauptthemen laut MEDLINE: Emergency Service, Hospital, Machine Learning, Blood Culture, Sepsis. Die Arbeit nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Western Australian Future Health and Innovation Fund, Government of Western Australia.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Bloodstream infections (BSIs) are particularly problematic in the emergency department (ED) of hospitals, where patients often present with undiagnosed illness, one cause of which may be undetected BSI. Identifying whether a patient needs a blood culture (BC) performed is one component of this challenge with implications for diagnostic efficiency and avoidance of unnecessary resource expenditure. Gap statement. In Western Australia, there has been no previous study investigating methods to predict BC outcome in an ED patient cohort.

Aim

This article assesses the feasibility of previously developed machine learning (ML) models for BC outcome prediction using a prospectively collected ED patient dataset. Methodology. An ML pipeline containing models previously trained using complete blood count (CBC), white blood cell differential (DIFF) and cell population data (CPD) generated by Sysmex XN-2000 haematology analysers was further evaluated using prospectively collected data containing patient sample results from the ED at Sir Charles Gairdner Hospital (SCGH), Perth, Western Australia. Blood samples used to produce CBC, DIFF and CPD were obtained at the same time as BC samples.

Results

There were 59 samples from 58 unique patients. Forty-nine of those samples were associated with negative BC results and 10 with positive BC results. We evaluated previously developed XGBoost (XG) and random forest (RF) ML models for positive BC outcome prediction. The RF and XG models obtained mean area under the receiver operating characteristic curve scores of 0.865 (95% CI, 0.763–0.947) and 0.833 (95% CI, 0.683–0.953) with the ED dataset.

Conclusion

The results presented in this study provide a foundation for further validation and shadow deployment of BC outcome prediction models in clinical settings and support future planning of clinical trials.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Emergency Service, HospitalMachine LearningBlood CultureSepsisBacteremiaHumansProspective StudiesFemalePredictive Learning ModelsMaleRandom ForestWestern AustraliaBoosting Machine Learning AlgorithmsMiddle AgedPrediction AlgorithmsAgedAdultClassification Algorithms

Stichwörter der Autor:innen

blood culturesbloodstream infectionmachine learning

Autor:innen und Institutionen

  • School of Physics, Mathematics and Computing, University of Western Australia, Perth 6009, Australia.
  • Discipline of Emergency Medicine, School of Medicine, University of Western Australia, Perth 6009, Australia.
  • Emergency Department, Sir Charles Gairdner Hospital, Perth 6009, Australia.
  • Discipline of Pathology and Laboratory Medicine, School of Medicine, University of Western Australia, Perth 6009, Australia.
  • Department of Microbiology, Pathwest Laboratory Medicine, Perth 6009, Australia.
  • Western Australian Country Health Service, Perth 6000, Australia.

Förderung

  • Western Australian Future Health and Innovation Fund, Government of Western Australia

Zitieren

APA

McFadden, B. R., Reynolds, M., Celenza, A., & Inglis, T. J. J. (2026). Evaluation of machine learning pipeline for blood culture outcome prediction on prospectively collected emergency department data. Journal of Medical Microbiology, 75(7). https://doi.org/10.1099/jmm.0.002191

Vancouver

McFadden BR, Reynolds M, Celenza A, Inglis TJJ. Evaluation of machine learning pipeline for blood culture outcome prediction on prospectively collected emergency department data. Journal of Medical Microbiology. 2026;75(7). doi:10.1099/jmm.0.002191

BibTeX
@article{McFadden2026,
  author = {McFadden, Benjamin R and Reynolds, Mark and Celenza, Antonio and Inglis, Timothy J J},
  title = {Evaluation of machine learning pipeline for blood culture outcome prediction on prospectively collected emergency department data},
  journal = {Journal of Medical Microbiology},
  year = {2026},
  volume = {75},
  number = {7},
  doi = {10.1099/jmm.0.002191}
}
Kostenlos starten →