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: Jul 7, 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

Combining machine learning and physiological network models for sepsis prediction.

Juri Backes¹, Artyom Tsanda^1,2, Tobias Knopp^1,2,3

¹Institute for Biomedical Imaging, Hamburg University of Technology, Hamburg, Germany.

Frontiers in Network Physiology

|July 6, 2026

Summary

Related Concept Videos

Model Approaches for Pharmacokinetic Data: Physiological Models

Model Approaches for Pharmacokinetic Data: Physiological Models

Physiological models in pharmacokinetics are instrumental in understanding the distribution and elimination of drugs within the body. These models describe the drug concentration within target organs, influenced by factors such as drug uptake, tissue volume, and blood flow. Drug uptake is governed by the partition coefficient, which signifies the drug concentration ratio in tissue to that in the blood. The blood flow rate to a specific tissue is expressed as Qt, and the rate of change in tissue...

You might also read

Related Articles

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

Sort by

Same author

Genetically engineered bacterial magnetosomes as optimized tracers for magnetic particle imaging.

Acta biomaterialia·2026

Same author

Deep learning for restoring MPI system matrices using simulated training data.

Physics in medicine and biology·2026

Same author

Continuous Coagulation Monitoring in Human Blood Samples via Magnetic Particle Spectroscopy.

International journal of nanomedicine·2026

Same author

The influence of muscle shapes on HDsEMG decomposition yield and accuracy.

Journal of electromyography and kinesiology : official journal of the International Society of Electrophysiological Kinesiology·2026

Same author

NMR relaxometry probes solvent-polarity-dependent molecular interactions in stimuli-responsive lyogels.

Physical chemistry chemical physics : PCCP·2025

Same author

A systematic delayed feedback control approach for unstable periodic orbits in chaotic systems with unknown parameters.

Chaos (Woodbury, N.Y.)·2025

This study introduces a new hybrid machine learning model for early sepsis prediction. The Latent Dynamics Model offers interpretable insights into patient health states, improving early detection of infectious disease complications.

Area of Science:

Computational Biology
Medical Informatics
Machine Learning

Background:

Sepsis is a life-threatening infectious disease with high mortality and long-term consequences.
Early prediction of sepsis onset is challenging due to complex pathophysiology.
Existing models offer either performance or interpretability, not both.

Purpose of the Study:

To develop a hybrid machine learning approach for interpretable early sepsis prediction.
To integrate a functional model of physiological interactions into a deep learning framework.
To provide clinically meaningful insights beyond standard risk scores.

Main Methods:

Proposed the Latent Dynamics Model, a hybrid approach combining coupled oscillators with machine learning.

Keywords:

coupled oscillator dynamical systems electronic health records hybrid modeling network physiology sepsis onset prediction

Related Experiment Videos

Last Updated: Jul 7, 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

Integrated physiological network models within a deep learning architecture.

Trained and evaluated the model on retrospective intensive care unit patient data (MIMIC-IV cohort).

Main Results:

Achieved competitive Area Under the Receiver Operating Characteristic Curve (AUROC) and Area Under the Precision-Recall Curve (AUPRC) performance.
Demonstrated interpretable differentiation between healthy and pathological states via model parameters.
Learned trajectories showed clinically plausible patterns of patient health dynamics.

Conclusions:

The Latent Dynamics Model provides an interpretable latent structure for sepsis onset prediction.
Embedding physiological network models in deep learning enhances predictive performance and interpretability.
This hybrid approach offers a promising direction for early sepsis detection and management.