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 28, 2026

A Data-Driven Approach to Quantifying Immune States in Sepsis

A Data-Driven Approach to Quantifying Immune States in Sepsis

Published on: February 7, 2025

Machine learning predicts sepsis deterioration trajectories.

Rui Zhang¹, Fang Long², Zhanqi Zhao^3,4

¹Department of Critical Care Medicine, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China. ccmzhangrui@foxmail.com.

NPJ Digital Medicine

|March 27, 2026

Summary

Related Concept Videos

Steps in Outbreak Investigation

Steps in Outbreak Investigation

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:

You might also read

Related Articles

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

Sort by

Same author

Genomic Insights into the Transmission Dynamics and Virulence of Carbapenem-Resistant Klebsiella pneumoniae in the ICU: A Retrospective Cohort Study.

Journal of global antimicrobial resistance·2026

Same author

Development and external validation of a prediction model for major adverse kidney events within 30 days in sepsis associated acute kidney injury: a multi-center retrospective clinical study.

BMC medical informatics and decision making·2026

Same author

H3K18la- driven neutrophil secretory autophagy promotes pulmonary endothelial dysfunction in sepsis-induced lung injury.

Autophagy·2026

Same author

Effects of inhaled nitric oxide on ventilation/perfusion mismatch assessed by electrical impedance tomography in intubated patients with moderate-to-severe ARDS: a prospective physiological study.

Respiratory research·2026

Same author

Electrical impedance tomography-derived flow index during spontaneous breathing trial stratifies the risk of reintubation within 48 h after extubation.

Critical care (London, England)·2026

Same author

Copper-Catalyzed Dehalogenative Cyclization/Sulfonylation of Alkynes: Synthesis of Pyrrolidone-Based Vinyl Sulfones.

The Journal of organic chemistry·2026

Same journal

Artificial intelligence for scoliosis surgical planning and postoperative prediction.

NPJ digital medicine·2026

Same journal

Enhancing anatomical recognition by surgeons during pelvic lymph node dissection using artificial intelligence.

NPJ digital medicine·2026

Same journal

AFP assistant: a retrieval-augmented generation and large language model-powered multilingual polio chatbot for low-resource language communities.

NPJ digital medicine·2026

Same journal

Structured reasoning failures compromise LLM interpretation of clinical oncology notes.

NPJ digital medicine·2026

Same journal

Translation of frozen sections into FFPE images for skin cancer resection margins using generative AI.

NPJ digital medicine·2026

Same journal

FedFound: a federated foundation model for lifespan brain morphological connectome analysis.

NPJ digital medicine·2026

See all related articles

This summary is machine-generated.

This study developed a dynamic sepsis prediction model using machine learning to identify patient trajectories. Early risk stratification using this model improved patient outcomes and reduced intensive care unit stays.

Area of Science:

Critical Care Medicine
Data Science in Healthcare
Computational Biology

Background:

Sepsis presents diverse clinical paths, but current scoring systems provide static risk assessments.
Dynamic, timely predictions are crucial for tailoring interventions in sepsis management.

Purpose of the Study:

To develop and validate a machine learning model for dynamic risk stratification in sepsis patients.
To identify distinct patient recovery trajectories in the intensive care unit (ICU).

Main Methods:

Group-based trajectory modeling was used on a large multicenter dataset (47,936 ICU patients).
An ensemble machine learning model incorporated dynamic physiological variability for prediction.
The model underwent temporal validation and external testing on public datasets (MIMIC-III, eICU, MIMIC-IV).

Related Experiment Videos

Last Updated: Mar 28, 2026

A Data-Driven Approach to Quantifying Immune States in Sepsis

A Data-Driven Approach to Quantifying Immune States in Sepsis

Published on: February 7, 2025

Main Results:

Three distinct sepsis trajectories were identified: rapid recovery (41.5%), slow recovery (36.4%), and clinical deterioration (22.1%).
The model achieved high predictive accuracy (AUROC 0.77-0.92) with a median warning time of 17.6 hours before deterioration.
Reduced heart rate variability was a significant predictor of mortality (adjusted HR 2.17).

Conclusions:

The validated trajectory-based model enables accurate, early sepsis risk stratification.
Implementation led to reduced ICU length of stay, mechanical ventilation duration, and 28-day mortality.
This approach supports proactive, individualized critical care for sepsis patients.