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

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

PDA (Privacy-Preserving Distributed Algorithms) in action: ten principles for high-quality multi-site clinical evidence generation.

Journal of the American Medical Informatics Association : JAMIA·2026

Same author

Promoting Problem-Solving Among Low-Income Adults With Type 2 Diabetes: Cluster-Randomized Controlled Trial of a Mobile Health Intervention With SMS Text Messaging (Mobile Diabetes Detective).

Journal of medical Internet research·2026

Same author

Real-world performance of large-scale propensity score adjustment strategies: Matching, weighting, and stratification.

Research square·2026

Same author

A comparison of Fast Healthcare Interoperability Resources and Observational Medical Mutcomes Partnership electronic health record data within the All of Us Research Program.

Journal of the American Medical Informatics Association : JAMIA·2026

Same author

Determinants of Delayed Recovery of Consciousness After Analgosedation Discontinuation in the ICU: Insights From Patients With COVID-19 Hypoxemic Respiratory Failure.

Critical care medicine·2026

Same author

Heterogeneity of Treatment Effects Across Nine Glucose-Lowering Drug Classes in Type 2 Diabetes: Extension of the LEGEND-T2DM Network Study.

Diabetes, obesity & metabolism·2026

Search research articles

Related Experiment Video

Updated: Jul 18, 2025

A Detailed Protocol for Physiological Parameters Acquisition and Analysis in Neurosurgical Critical Patients

A Detailed Protocol for Physiological Parameters Acquisition and Analysis in Neurosurgical Critical Patients

Published on: October 17, 2017

Interpretable physiological forecasting in the ICU using constrained data assimilation and electronic health record

David Albers¹, Melike Sirlanci², Matthew Levine³

¹Department of Biomedical Informatics, University of Colorado Anschutz Medical Campus, Aurora, 80045, CO, USA; Department of Biomedical Engineering, University of Colorado Anschutz Medical Campus, Aurora, 80045, CO, USA; Department of Biostatistics and Informatics, University of Colorado Anschutz Medical Campus, Aurora, 80045, CO, USA; Department of Biomedical Informatics, Columbia University, New York, 10032, NY, USA.

Journal of Biomedical Informatics

|August 21, 2023

Summary

Data assimilation (DA) with constrained inference improves physiological forecasting using sparse electronic health record data. This method enhances accuracy and reduces data needs for critical care predictions.

Keywords:

Constrained ensemble Kalman filter Data assimilation Electronic health record data Glucose–insulin Mathematical physiological model

More Related Videos

Monitoring Lung Function with Electrical Impedance Tomography in the Intensive Care Unit

Monitoring Lung Function with Electrical Impedance Tomography in the Intensive Care Unit

Published on: September 6, 2024

Real-Time Monitoring of Neurocritical Patients with Diffuse Optical Spectroscopies

Real-Time Monitoring of Neurocritical Patients with Diffuse Optical Spectroscopies

Published on: November 19, 2020

Related Experiment Videos

Last Updated: Jul 18, 2025

A Detailed Protocol for Physiological Parameters Acquisition and Analysis in Neurosurgical Critical Patients

A Detailed Protocol for Physiological Parameters Acquisition and Analysis in Neurosurgical Critical Patients

Published on: October 17, 2017

Monitoring Lung Function with Electrical Impedance Tomography in the Intensive Care Unit

Monitoring Lung Function with Electrical Impedance Tomography in the Intensive Care Unit

Published on: September 6, 2024

Real-Time Monitoring of Neurocritical Patients with Diffuse Optical Spectroscopies

Real-Time Monitoring of Neurocritical Patients with Diffuse Optical Spectroscopies

Published on: November 19, 2020

Area of Science:

Computational biology
Physiological modeling
Data science in medicine

Background:

Physiological prediction is crucial for medical interventions but challenging due to complex, sparse, and non-stationary clinical data.
Traditional computational methods struggle with the noise and limited nature of electronic health record (EHR) data.
Data assimilation (DA) offers a promising approach by integrating mechanistic models with inference techniques.

Purpose of the Study:

To develop and evaluate a novel data assimilation framework for predicting physiological processes using sparse EHR data.
To forecast blood glucose levels in intensive care unit (ICU) patients by embedding a glucose-insulin model into a new DA framework.
To assess the impact of constrained inference on forecasting accuracy and data requirements.

Main Methods:

Developed a methodological pipeline using a constrained ensemble Kalman filter (CEnKF) to forecast blood glucose.
Extracted complex patient data, including nutrition, glucose, insulin, and medications, manually for accuracy.
Estimated models in real-time, compared constrained and unconstrained filters, and varied model parameters and constraints to assess forecasting accuracy.

Main Results:

The novel CEnKF demonstrated significant improvements in robustness and accuracy, requiring less data than unconstrained filters.
Accurate forecasting depended on administered insulin and tube nutrition; IV glucose delivery did not enhance accuracy.
Model flexibility, influenced by constraint boundaries and parameter estimation, impacted forecasting performance.

Conclusions:

Accurate and robust physiological forecasting is achievable with DA, even with sparse clinical data.
Constrained inference, especially on unmeasured states and parameters, effectively reduces forecast error and data demands.
While model flexibility is important, overly strict or loose constraints can increase forecasting errors.