Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jul 21, 2026

Evaluation of a Reliable Biomarker in a Cecal Ligation and Puncture-Induced Mouse Model of Sepsis

Evaluation of a Reliable Biomarker in a Cecal Ligation and Puncture-Induced Mouse Model of Sepsis

Published on: December 9, 2022

Service-Specific Heterogeneity in Sepsis Variable Significance and Machine Learning Model Performance: A Stratified

Marcio Borges-Sa^1,2,3,4, Eric Macias-Fassio^5,6, Alejandro Delgado⁵

¹Multidisciplinary Sepsis Unit, Son Llatzer University Hospital, 07198 Palma de Mallorca, Spain.

Journal of Clinical Medicine

|July 15, 2026

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

The ICU pharmacist's contribution to antimicrobial optimization through routine pharmacist interventions.

Revista espanola de quimioterapia : publicacion oficial de la Sociedad Espanola de Quimioterapia·2026

Same author

From development to clinical practice: deployment of an interoperable and secure ML-based CDSS to aid in the early detection of sepsis.

International journal of medical informatics·2026

Same author

Diagnostic accuracy of PCT, IL-6, and MR-ProADM for early identification of sepsis in the emergency department.

Revista espanola de quimioterapia : publicacion oficial de la Sociedad Espanola de Quimioterapia·2026

Same author

Privacy-Aware Synthetic Tabular Data Generation for Healthcare: Application to Sepsis Detection.

Bioengineering (Basel, Switzerland)·2026

Same author

Prognostic Value of International Normalized Ratio and Thrombocytopenia in Early Risk Stratification of Septic Patients.

Biomedicines·2026

Same author

Comment on: Is "pre-sepsis" the new sepsis? A narrative review.

PLoS pathogens·2026

Sepsis detection varies by hospital setting. Machine learning models, especially BIAlert, outperform traditional scoring systems by maintaining sepsis detection accuracy across diverse clinical environments.

Area of Science:

Medical informatics
Clinical decision support systems
Sepsis research

Background:

Sepsis detection relies on clinical variables and scoring systems, often assuming uniform performance across hospital settings.
Sepsis phenotype distributions vary between clinical environments, impacting the reliability of standard detection methods.
Variable importance in sepsis detection may be setting-dependent, necessitating context-specific evaluation.

Purpose of the Study:

To quantify service-specific variability in the discriminatory capacity of clinical variables for sepsis detection.
To evaluate if heterogeneity in variable importance translates to differential performance of machine learning (ML) models versus traditional clinical scoring systems.
To assess the robustness of ML models and clinical rules across different hospital environments.

Keywords:

Random Forest artificial intelligence biomarkers clinical decision support clinical heterogeneity hospital department machine learning scoring systems sepsis sepsis phenotypes

More Related Videos

A Data-Driven Approach to Quantifying Immune States in Sepsis

A Data-Driven Approach to Quantifying Immune States in Sepsis

Published on: February 7, 2025

Related Experiment Videos

Last Updated: Jul 21, 2026

Evaluation of a Reliable Biomarker in a Cecal Ligation and Puncture-Induced Mouse Model of Sepsis

Evaluation of a Reliable Biomarker in a Cecal Ligation and Puncture-Induced Mouse Model of Sepsis

Published on: December 9, 2022

A Data-Driven Approach to Quantifying Immune States in Sepsis

A Data-Driven Approach to Quantifying Immune States in Sepsis

Published on: February 7, 2025

Main Methods:

Stratified sub-analysis of the BIAlert Sepsis cohort (203,755 patients, 11,864 sepsis episodes, 2014-2018).
Evaluation of 61 structured quantitative variables across nine hospital services using Mann-Whitney-Wilcoxon tests.
Comparison of five ML models (including BIAlert) and three clinical rules (NEWS, SIRS, qSOFA) globally and stratified by clinical environment.

Main Results:

Variable significance for sepsis detection ranged from 95.1% in the Emergency Department to 37.7% in the Intensive Care Unit.
Lactate was the only universally significant variable; traditional scoring systems (qSOFA, NEWS) showed poor performance in Critical Care (AUC 0.459).
The BIAlert ML model maintained the highest Area Under the Curve (AUC) across all environments (0.975-0.857), outperforming other models and rules.

Conclusions:

The discriminatory capacity of clinical variables for sepsis detection is not uniform across hospital services.
Machine learning models, particularly BIAlert, demonstrate robust and consistent performance in sepsis detection.
ML models maintain effectiveness in diverse clinical settings where fixed-rule scoring systems may fail.