Early Identification of Recovery Potential After Acute Brain Injury Using Functional Near-Infrared Spectroscopy
Matthew Kolisnyk1, Karnig Kazazian2,3, Garima Gupta2,3
1Western Institute of Neuroscience, Western University, London, ON, Canada. mkolisny@uwo.ca.
Neurocritical Care
|July 24, 2026
Summary
Functional near-infrared spectroscopy (fNIRS) can predict long-term outcomes in acute brain injury patients. This bedside tool offers objective neural measures, outperforming traditional assessments for early prognostication.
Area of Science:
- Neuroscience
- Medical Technology
- Critical Care Medicine
Background:
- Early prognostication in acute brain injury is challenging.
- Conventional assessments offer limited insight into long-term outcomes.
- Functional neuroimaging is underutilized in intensive care due to cost and logistics.
Purpose of the Study:
- To evaluate functional near-infrared spectroscopy (fNIRS) for early prognostication in acute brain injury.
- To assess fNIRS's ability to predict 6-month functional outcomes.
- To compare fNIRS performance with clinical and behavioral assessments.
Main Methods:
- Prospective observational cohort study of 33 acute brain injury patients.
- fNIRS recordings while patients listened to audio movie clips.
- Machine learning model trained on functional connectivity features to predict outcomes.
Main Results:
- fNIRS model predicted 6-month outcome with 81.3% balanced accuracy, outperforming clinical models.
- The model also predicted behavioral responsiveness recovery.
- fNIRS did not accurately predict covert awareness.
Conclusions:
- Bedside fNIRS provides objective neural measures for functional recovery prediction in acute brain injury.
- fNIRS may capture clinically relevant brain function missed by conventional methods.
- Further validation in larger cohorts could establish fNIRS as a complementary prognostication tool.
