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 Experiment Video

Updated: May 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

598

Predicting sepsis prognosis using deep learning with routine biomarkers.

Qi Yun Gan1, Xin Li2, Yuan Xue3

  • 1Department of Emergency Medicine, Liuzhou People's Hospital Affiliated to Guangxi Medical University, Liuzhou, China.

Frontiers in Medicine
|May 4, 2026
PubMed
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

Partial resuscitative endovascular balloon occlusion of the aorta as a resuscitative strategy in traumatic hemorrhagic shock: A retrospective analysis of a Chinese single-center cohort.

Chinese journal of traumatology = Zhonghua chuang shang za zhi·2026
Same author

Resuscitative endovascular balloon occlusion of the aorta (REBOA) in interhospital trauma care: A case report enabled by real-time information coordination.

Medicine·2026
Same author

Analysis of the effects of early screening combined with blood lactate on the severity of patients with sepsis.

Heliyon·2024
Same author

Identification of PANoptosis-related subtypes, construction of a prognosis signature, and tumor microenvironment landscape of hepatocellular carcinoma using bioinformatic analysis and experimental verification.

Frontiers in immunology·2024
Same author

Ampelopsin attenuates Staphylococcus aureus Alpha-Toxin-Induced Lung Injury.

Microbial pathogenesis·2023
Same author

[Effects of beta-catenin-specific siRNA interference on Jurkat and K562 cells].

Zhongguo yi xue ke xue yuan xue bao. Acta Academiae Medicinae Sinicae·2008

This study developed an AI model using red cell distribution width (RDW) and estimated plasma volume status (ePVS) to predict sepsis mortality. The model showed high accuracy in predicting 28-day mortality in emergency sepsis patients.

Area of Science:

  • Critical care medicine
  • Artificial intelligence in healthcare
  • Biomarker discovery

Background:

  • Sepsis is a life-threatening condition requiring accurate early prognosis for effective treatment.
  • Emergency sepsis patients face rapid progression and complex changes, necessitating precise risk stratification.
  • Traditional biomarkers and scoring systems have limitations in predicting sepsis outcomes.

Purpose of the Study:

  • To develop an artificial intelligence (AI) model integrating red cell distribution width (RDW) and estimated plasma volume status (ePVS).
  • To predict the 28-day mortality risk in emergency department sepsis patients.
  • To enhance the accuracy of prognostic prediction beyond single biomarkers.

Main Methods:

  • A multilayer perceptron (MLP) deep learning model was constructed using data from 73 sepsis patients.
Keywords:
artificial neural networkdeep learningestimated plasma volume statusprognosis predictionred blood cell distribution widthsepsis

Related Experiment Videos

Last Updated: May 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

598
  • Internal validation was performed on the initial dataset.
  • External validation utilized independent MIMIC-III and MIMIC-IV datasets for robust performance evaluation.

    • Main Results:

    • The deep learning model achieved 87.0% accuracy internally and 92.0% externally.
    • External validation showed 84.0% sensitivity and 83.0% specificity.
    • The combined RDW and ePVS model achieved an AUC of 0.812, outperforming individual indicators.

    Conclusions:

    • A deep learning model integrating RDW and ePVS significantly improves prognostic prediction accuracy for emergency sepsis.
    • Elevated RDW and ePVS are potential indicators of adverse outcomes in sepsis patients.
    • This AI-driven approach offers a valuable tool for early risk identification and clinical decision support.