Related Experiment Video
Updated: Mar 19, 2026
A Detailed Protocol for Physiological Parameters Acquisition and Analysis in Neurosurgical Critical Patients
Published on: October 17, 2017
Predicting Intracranial Pressure and Brain Tissue Oxygen Crises in Patients With Severe Traumatic Brain Injury
Risa B Myers1, Christos Lazaridis, Christopher M Jermaine
11Department of Computer Science, Rice University, Houston, TX.2Division of Neurocritical Care, Department of Neurology, Baylor College of Medicine, Houston, TX.3Department of Neurosurgery, Baylor College of Medicine, Houston, TX.4Department of Pediatrics, Section of Cardiology, Baylor College of Medicine, Houston, TX.
Computer algorithms can predict intracranial pressure and brain tissue oxygenation crises in traumatic brain injury patients, offering crucial advance warning for timely clinical intervention and improved outcomes.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Critical Care Medicine
Background:
- Traumatic brain injury (TBI) management requires continuous monitoring for secondary injury.
- Intracranial pressure (ICP) and brain tissue oxygenation (ptiO2) crises are critical events in TBI.
- Early detection of these crises is vital for effective clinical intervention.
Purpose of the Study:
- To develop predictive algorithms for intracranial pressure (ICP) and partial brain tissue oxygenation (ptiO2) crises in TBI patients.
- To identify physiological patterns preceding crisis onset for early detection.
- To enable timely interventions to prevent or mitigate secondary brain injury.
Main Methods:
- Retrospective analysis of prospectively collected physiological data from 817 severe TBI patients.
- Defined ICP crises (≥20 mmHg for ≥15 min) and ptiO2 crises (<10 mmHg for ≥10 min).
- Applied multivariate classification models to 30-minute epochs to predict crises 15-360 minutes in advance.
Main Results:
- Algorithm predicted ICP crises with 30-minute warning (AUC 0.86) using ICP and time since last crisis.
- Algorithm predicted ptiO2 crises with 30-minute warning (AUC 0.91).
- High predictive accuracy achieved using readily available monitoring data.
Conclusions:
- Developed algorithms accurately predict ICP and ptiO2 crises in severe TBI.
- These algorithms offer timely, 30-minute advance warnings.
- Prediction relies primarily on the signal of interest and time since the last crisis, simplifying implementation.
