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: Dec 8, 2025

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

408

Automated prediction of sepsis using temporal convolutional network.

Christopher Kok1, V Jahmunah1, Shu Lih Oh1

  • 1School of Engineering, Ngee Ann Polytechnic, Singapore.

Computers in Biology and Medicine
|September 17, 2020
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

Spatiotemporal deep learning for early detection of isolated REM sleep behavior disorder and Parkinson's disease using functional MRI data.

NPJ Parkinson's disease·2026
Same author

Current approaches to the diagnosis and management of retained products of conception and enhanced myometrial vascularity of the first trimester.

The European journal of contraception & reproductive health care : the official journal of the European Society of Contraception·2026
Same author

Anti-Dengue IgG Seroprevalence and Exposure-Related Risk in Italian Military Personnel Deployed on Overseas Missions: A Cross-Sectional Study.

Tropical medicine and infectious disease·2026
Same author

An integrated calibration-uncertainty framework for improving the reliability of deep learning models for seizure detection in EEG signals.

Computer methods and programs in biomedicine·2026
Same author

Why almost all ML models for medicine are wrong-and what we need for evidence-based medical AI.

International journal of medical informatics·2026
Same author

Impact of Pollution on Mental Health: A Systematic Review of Associations, Methodological Challenges, and Future Directions.

Health science reports·2026

A novel deep learning model accurately predicts sepsis, a life-threatening condition. This automated diagnostic tool achieved high accuracy in predicting sepsis events and patient outcomes, offering a promising advancement for early detection in hospitals.

Area of Science:

  • Medical Informatics
  • Artificial Intelligence in Medicine
  • Critical Care Medicine

Background:

  • Sepsis is a life-threatening condition characterized by multiple organ failure due to the body's dysregulated response to infection.
  • Current diagnostic methods for sepsis face limitations, necessitating the development of advanced tools for timely and accurate detection.
  • Early identification of sepsis is crucial for effective treatment and improved patient outcomes, reducing mortality and morbidity.

Purpose of the Study:

  • To develop and validate a cost-effective, automated diagnostic tool for the prediction of sepsis.
  • To investigate the efficacy of a deep temporal convolution network for sepsis prediction using both per time-step and per-patient metrics.
  • To establish a robust and precise model for early sepsis detection in clinical settings.
Keywords:
10-Fold validationDeep learningMachine learningPredictionSepsisTemporal convolution networkper time-step metricsper-patient metrics

Related Experiment Videos

Last Updated: Dec 8, 2025

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

408

Main Methods:

  • Development of a deep temporal convolution network (DTCN) model for sepsis prediction.
  • Training and evaluation of the DTCN model using septic patient data.
  • Assessment of model performance using per time-step and per-patient metrics, including accuracy and Area Under the Receiver Operating Characteristic Curve (AUROC).
  • Validation of the model through three distinct validation methods.

Main Results:

  • The DTCN model achieved high performance metrics for sepsis prediction.
  • Per time-step metrics demonstrated an accuracy of 98.8% and an AUROC of 98.0%.
  • Per-patient metrics showed a high accuracy of 95.5% and an AUROC of 91.0%.
  • The study uniquely investigated per time-step metrics, differentiating it from prior research focused on per-patient metrics.

Conclusions:

  • The developed deep temporal convolution network model is a robust and accurate tool for sepsis prediction.
  • The model's high precision and validated performance indicate its potential for clinical implementation in hospitals.
  • This automated diagnostic approach offers a significant advancement in the early detection and management of sepsis.