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

Steps in Outbreak Investigation01:18

Steps in Outbreak Investigation

459
In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis:
459

You might also read

Related Articles

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

Sort by
Same author

Dynamic changes in mean corpuscular volume could be integrated into longitudinal prediction frameworks for individualized risk assessment in the ICU setting: A retrospective study.

Science progress·2026
Same author

Progranulin Is a Useful Biomarker to Predict Mortality in ICU Patients with Low Burden of Organ Dysfunction.

Biomedicines·2026
Same author

Interleukin-6 in Daily Use in the Intensive Care Unit: Does It Change the Patients' Outcome and Antimicrobial Prescription? An Explorative Study.

Life (Basel, Switzerland)·2026
Same author

dsLassoCov: a federated Lasso approach incorporating covariate control.

Scientific reports·2026
Same author

A multi-task learning approach combining regression and classification tasks for joint feature selection.

Scientific reports·2026
Same author

Comparative assessment of extravascular lung water index in ARDS patients on veno-venous ECMO: transpulmonary thermodilution versus AI-driven chest CT segmentation.

BMC pulmonary medicine·2026

Related Experiment Video

Updated: Jan 5, 2026

A Data-Driven Approach to Quantifying Immune States in Sepsis
07:42

A Data-Driven Approach to Quantifying Immune States in Sepsis

Published on: February 7, 2025

452

Leveraging implicit expert knowledge for non-circular machine learning in sepsis prediction.

Shigehiko Schamoni1, Holger A Lindner2, Verena Schneider-Lindner3

  • 1Department of Computational Linguistics, Heidelberg University, Germany; Interdisciplinary Center for Scientific Computing, Heidelberg University, Germany.

Artificial Intelligence in Medicine
|October 15, 2019
PubMed
Summary
This summary is machine-generated.

Accurate sepsis prediction models are crucial for patient survival. This study introduces a novel method using physician judgments to create independent sepsis labels, achieving state-of-the-art results and improving model validity.

Keywords:
Machine learning in health careSepsis prediction

More Related Videos

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

8.6K

Related Experiment Videos

Last Updated: Jan 5, 2026

A Data-Driven Approach to Quantifying Immune States in Sepsis
07:42

A Data-Driven Approach to Quantifying Immune States in Sepsis

Published on: February 7, 2025

452
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

8.6K

Area of Science:

  • Critical Care Medicine
  • Machine Learning in Healthcare
  • Clinical Informatics

Background:

  • Sepsis is a leading cause of mortality in intensive care units, with delayed antibiotic treatment significantly increasing mortality risk.
  • Accurate early sepsis prediction models are vital for timely intervention and improved patient outcomes.
  • Existing machine learning models often use automatically defined sepsis labels derived from clinical criteria, creating a circular dependency that may compromise model validity.

Purpose of the Study:

  • To develop a more valid approach for creating ground truth sepsis labels for machine learning research.
  • To investigate the effectiveness of using implicit clinical practitioner knowledge for sepsis identification.
  • To build and evaluate machine learning models for early sepsis prediction using an independent ground truth dataset.

Main Methods:

  • Developed an electronic questionnaire to capture attending physicians' daily judgments of patient sepsis status, creating an independent ground truth dataset.
  • Trained machine learning models using this novel dataset, contrasting with previous methods relying on automatically derived labels.
  • Analyzed feature contributions in a linear model to understand surprising or counterintuitive findings.

Main Results:

  • Despite a small dataset size, the proposed method achieved state-of-the-art Area Under the Receiver Operating Characteristic (AUROC) scores.
  • Inspection of learned model weights revealed potentially unexpected feature contributions, offering insights into sepsis prediction.
  • The approach demonstrated the feasibility of using physician-based judgments for robust sepsis research.

Conclusions:

  • Creating independent ground truth labels by leveraging clinical expertise offers a more valid foundation for sepsis prediction models.
  • This method overcomes the inherent circularity of using clinical criteria as both labels and features.
  • The findings support the development of more reliable machine learning tools for early sepsis detection in clinical practice.