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: Mar 18, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Machine Learning in the ICU: Predicting Mortality in Patients with Carbapenem-Resistant Gram-Negative Bacilli

Özlem Güler¹, Volkan Alparslan², Burak İnner³

¹Department of Infectious Diseases and Clinical Microbiology, Kocaeli University Medical School, Izmit, Turkey.

Journal of Intensive Care Medicine

|March 16, 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

A Hybrid CNN-MLLM Architecture for Image-Based Nutrition Estimation and Advisory Insulin Decision Support in Type 1 Diabetes.

Nutrients·2026

Same author

Diagnostic performance comparison of multiplex PCR-based BioFire® joint infection panel and conventional culture methods in septic arthritis: A single center, retrospective study.

Diagnostic microbiology and infectious disease·2026

Same author

Tuberculosis: A Century of Struggle and an Ongoing Global Challenge.

Infectious diseases & clinical microbiology·2026

Same author

Brain volumetric changes in pediatric seronegative autoimmune encephalitis: A case-control study.

Journal of neuroimmunology·2026

Same author

Effects of air pollution and meteorological factors on hematological exacerbation phenotypes in patients presenting to the emergency department with COPD exacerbation.

International journal of biometeorology·2026

Same author

Early Identification of Atherosclerosis in People Living with HIV by Coronary Computed Tomography Angiography.

Diagnostics (Basel, Switzerland)·2026

A machine learning model accurately predicts mortality in intensive care unit (ICU) patients with multidrug-resistant gram-negative bacilli infections. This tool aids early intervention and end-of-life care decisions for these critical infections.

Area of Science:

Critical care medicine
Infectious diseases
Machine learning in healthcare

Background:

Bloodstream infections caused by multidrug and carbapenem-resistant gram-negative bacilli (MDR/CRGNB) are associated with high mortality in intensive care units (ICUs).
Accurate mortality prediction is crucial for timely treatment adjustments and informed end-of-life care planning.

Purpose of the Study:

To develop and validate a machine learning model for predicting 15-day mortality in ICU patients diagnosed with MDR/CRGNB bloodstream infections.
To identify key clinical and laboratory factors influencing mortality in this patient population.

Main Methods:

A retrospective cohort study involving 197 adult ICU patients with MDR/CRGNB bloodstream infections (Klebsiella pneumoniae, Pseudomonas aeruginosa, Acinetobacter baumannii) from 2017-2023.

Keywords:

antimicrobial drug resistance gram-Negative bacteria intensive care unit mortality supervised-Machine learning

Related Experiment Videos

Last Updated: Mar 18, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Evaluation of ten machine learning classifiers using 5-fold cross-validation, with performance metrics including AUROC, AUPRC, accuracy, precision, recall, F1 score, MCC, and Brier score.

Interpretation of model predictions using SHapley Additive exPlanations (SHAP) to identify significant risk factors.

Main Results:

The Light Gradient Boosting Machine (LightGBM) classifier achieved superior performance, demonstrating an AUROC of 0.94 and AUPRC of 0.952.
Key predictors of mortality identified by SHAP analysis included coagulopathy, rapid antibiotic administration, septic shock, SOFA score, platelet count, CRP level, and time-related variables.
The overall 15-day mortality rate in the cohort was 48%.

Conclusions:

The developed LightGBM model shows significant promise for accurately predicting mortality in ICU patients with MDR/CRGNB bloodstream infections.
This predictive tool can potentially facilitate earlier clinical interventions and support critical decision-making regarding patient care and end-of-life planning.