Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Pharmacokinetic Models: Comparison and Selection Criterion01:26

Pharmacokinetic Models: Comparison and Selection Criterion

490
Physiological and compartmental models are valuable tools used in studying biological systems. These models rely on differential equations to maintain mass balance within the system, ensuring an accurate representation of the dynamic processes at play.
Physiological models take a detailed approach by considering specific molecular processes. They can predict drug distribution, metabolism, and elimination changes, providing a comprehensive understanding of how drugs interact with the body.
490
Receiver Operating Characteristic Plot01:15

Receiver Operating Characteristic Plot

584
A ROC (Receiver Operating Characteristic) plot is a graphical tool used to assess the performance of a binary classification model by illustrating the trade-off between sensitivity (true positive rate) and specificity (false positive rate). By plotting sensitivity against 1 - specificity across various threshold settings, the ROC curve shows how well the model distinguishes between classes, with a curve closer to the top-left corner indicating a more accurate model. The area under the ROC curve...
584
Impact of Pharmacokinetic–Pharmacodynamic Models: Regulatory Decisions01:15

Impact of Pharmacokinetic–Pharmacodynamic Models: Regulatory Decisions

97
PK–PD modeling has significantly influenced FDA regulatory decisions, particularly drug approval, dosage optimization, and labeling. These models integrate pharmacokinetics (PK) and pharmacodynamics (PD) to predict drug behavior and effects, aiding in optimizing dosing regimens and enhancing the probability of clinical trial success.One notable example is Nesiritide (Natrecor®), a recombinant human brain natriuretic peptide for treating acute decompensated congestive heart failure...
97

You might also read

Related Articles

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

Sort by
Same author

<i>Letter:</i> Author Response to Letter to the Editor from Luo et al.

Respiratory care·2026
Same author

Post-Acute Ventilator Weaning and Discharge Outcomes in Individuals With Cervical Spinal Cord Injury.

Respiratory care·2026
Same author

Surgical and N95 Masks in Chronic Hypercapnic Respiratory Failure: What Physiology Can and Cannot Tell Us.

Respiratory care·2026
Same author

Scalable Agile Framework for Execution in AI for Medical AI Ethics Policy Design in Small- and Medium-Sized Enterprises.

Journal of medical Internet research·2026
Same author

Immersive Virtual Reality Use in Medical Intensive Care: Mixed Methods Feasibility Study.

JMIR serious games·2024
Same author

The association of weight loss from anti-obesity medications or bariatric surgery and apnea-hypopnea index in obstructive sleep apnea.

Obesity reviews : an official journal of the International Association for the Study of Obesity·2024

Related Experiment Video

Updated: May 5, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

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

Published on: October 10, 2018

8.0K

Decision Curve Analysis for Evaluating Machine Learning Models for Next-Day Transfer Out of ICU.

Margaret Pozo1,2, Abigail Pape3, Brian Locke4

  • 1Department of Biomedical Engineering, The University of Utah, USA.

Medrxiv : the Preprint Server for Health Sciences
|May 4, 2026
PubMed
Summary

Predicting intensive care unit (ICU) patient transfers using machine learning models can optimize workflows. Decision curve analysis helps determine the clinical utility of these predictions for proactive patient management.

Related Experiment Videos

Last Updated: May 5, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

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

Published on: October 10, 2018

8.0K

Area of Science:

  • Clinical Informatics
  • Machine Learning in Healthcare
  • Health Services Research

Background:

  • Timely identification of intensive care unit (ICU) patients nearing transfer is crucial for efficient healthcare operations.
  • Traditional prediction metrics like discrimination and calibration do not fully capture the practical implications of using predictive models in clinical settings.
  • Existing methods for predicting ICU discharge lack a clear framework for evaluating their real-world decision-making utility.

Purpose of the Study:

  • To develop and evaluate machine learning models for predicting next-day ICU transfers.
  • To assess the clinical utility and decision consequences of these predictions using decision curve analysis (DCA).
  • To translate model performance into practical operational workflows, considering real-world constraints like time limitations.

Main Methods:

  • Utilized adult ICU admission data from MIMIC-IV, representing patient stays as daily clinical summaries.
  • Trained and compared logistic regression, random forest, and XGBoost models to predict next-day ICU transfer.
  • Employed decision curve analysis (DCA) to evaluate the net benefit and clinical utility of model-guided prediction strategies across various thresholds.

Main Results:

  • Machine learning models achieved high discrimination (ROC AUC 0.80-0.84) with varying calibration.
  • Model-guided strategies demonstrated superior decision utility compared to 'review-all', 'review-none', or simple clinical rules across tested thresholds.
  • In a simulated clinical trial recruitment workflow, a feasible operating threshold (0.23) predicted approximately 1.23 enrollments per day, considering time constraints and conservative assumptions.

Conclusions:

  • Decision curve analysis offers a transparent method for assessing the value and optimal threshold selection for ICU transfer prediction models.
  • Machine learning models, when evaluated with DCA, provide actionable insights for improving proactive patient management and operational efficiency.
  • The study demonstrates a pathway to integrate predictive modeling into real-world clinical workflows, aligning predictions with operational constraints and maximizing net benefit.