EUVIMEDEUVIMED entdecken
Open Access

How Can Clinicians Decide if AI‐Enabled Risk Prediction Models Are Fit for Purpose for Real World Use?

Ian A Scott iD, Anton H van der Vegt, Victoria Campbell, Paul J Lane, Michael Rice, Balasubramanian Venkatesh

Journal of Evaluation in Clinical Practice · 2026

DOI: 10.1111/jep.70557 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Evaluation in Clinical Practice, 2026. Verfasst von Ian A Scott und 5 weiteren Autor:innen. Beteiligte Einrichtung: Digital Health and Informatics Directorate, Metro South Hospital and Health Service, Brisbane, Queensland, Australia. Hauptthemen laut MEDLINE: Artificial Intelligence, Sepsis, Clinical Decision-Making. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: Queensland Government, Clinical Excellence Queensland.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Artificial intelligence (AI)‐enabled risk prediction models are being increasingly used in hospital practice to predict patient risk of adverse events, aimed at giving early warning of impending events and facilitating preventive intervention. However, evidence of beneficial impact varies which may be due to clinicians not finding them useful because of suboptimal predictive accuracy (effectiveness), burden of false alerts (efficiency) or narrow prediction windows (utility). Current model performance measures do not capture the interdependency of these three dimensions. In this commentary, using sepsis risk prediction as an example, we present methods that assist clinicians to: assess the suitability of particular models and decide ideal decision thresholds; optimise model performance; and configure and deliver alerts to frontline clinicians.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Artificial IntelligenceSepsisClinical Decision-MakingHumansRisk AssessmentPrediction AlgorithmsPredictive Learning Models

Stichwörter der Autor:innen

effectivenessefficiencyhospitalmachine learningrisk prediction modelssepsisutility

Autor:innen und Institutionen

  • Digital Health and Informatics Directorate, Metro South Hospital and Health Service, Brisbane, Queensland, Australia.
  • Centre for Health Services Research, University of Queensland, Brisbane, Queensland, Australia.
  • Department of Intensive Care, Sunshine Coast University Hospital, Burtinya, Queensland, Australia.
  • Safety Quality and Innovation, The Prince Charles Hospital, Brisbane, Queensland, Australia.
  • Patient Safety and Quality, Clinical Excellence Queensland, Brisbane, Queensland, Australia.

ORCID-Profile: Ian A Scott

Förderung

  • Queensland Government
  • Clinical Excellence Queensland

Zitieren

APA

Scott, I. A., van der Vegt, A. H., Campbell, V., Lane, P. J., Rice, M., & Venkatesh, B. (2026). How Can Clinicians Decide if AI‐Enabled Risk Prediction Models Are Fit for Purpose for Real World Use?. Journal of Evaluation in Clinical Practice. https://doi.org/10.1111/jep.70557

Vancouver

Scott IA, van der Vegt AH, Campbell V, Lane PJ, Rice M, Venkatesh B. How Can Clinicians Decide if AI‐Enabled Risk Prediction Models Are Fit for Purpose for Real World Use?. Journal of Evaluation in Clinical Practice. 2026. doi:10.1111/jep.70557

BibTeX
@article{Scott2026,
  author = {Scott, Ian A and van der Vegt, Anton H and Campbell, Victoria and Lane, Paul J and Rice, Michael and Venkatesh, Balasubramanian},
  title = {How Can Clinicians Decide if AI‐Enabled Risk Prediction Models Are Fit for Purpose for Real World Use?},
  journal = {Journal of Evaluation in Clinical Practice},
  year = {2026},
  doi = {10.1111/jep.70557}
}
Kostenlos starten →