Two graduate students from Western University have developed a groundbreaking method for predicting brain injury recovery in ICU patients.
Matthew Kolisnyk and Karnig Kazazian, PhD candidates at the Schulich School of Medicine & Dentistry in the lab of neuroscientist Adrian Owen, combined functional magnetic resonance imaging (fMRI) with Machine Learning techniques to predict which patients will survive a severe brain injury.
The study, ‘Predicting neurologic recovery after severe acute brain injury using resting-state networks,’ is published in the Journal of Neurology.
Uncertainty around brain injury recovery
Whether a traumatic brain injury is caused by a stroke or cardiac arrest, lives can be changed forever.
When patients with serious brain injuries are admitted to the ICU, families are faced with great uncertainty around recovery.
Equally uncertain about brain injury recovery, however, are the health professionals.
“For years we’ve lacked the tools and techniques to know who is going to survive a serious brain injury,” said Owen.
Solving the problem of recovery uncertainty
An interdisciplinary team of researchers from Western, in collaboration with neurologists at the London Health Sciences Centre and Lawson Health Research Institute, aimed to combat the uncertainty around brain injury recovery.
The team was led by Loretta Norton, a psychology professor at King’s University College at Western. Norton was one of the first researchers in the world to measure brain activity in the ICU.
The researchers measured brain activity in 25 patients at one of London’s two ICUs in the first few days after a serious brain injury. They then tested whether it could predict who would survive and who would not.
“We previously found that information about the potential for recovery in these patients was captured in the way different brain regions communicate with each other,” said Norton.
“Intact communication between brain regions is an important factor for regaining consciousness.”
Improving prediction accuracy with AI
The team’s breakthrough occurred when they discovered that they could combine this imaging technique with an Artificial Intelligence application known as Machine Learning.
With this combination, they found that they could predict brain injury recovery with an accuracy of 80%. This is higher than the current standard of care.
“Modern Artificial Intelligence has shown incredible predictive capabilities. Combining this with our existing imaging techniques was enough to better predict who will recover from their injuries,” said Kolisnyk.
More work must be done on the recovery prediction model
Although the results from the study are encouraging, the researchers say the prediction was not perfect and needs further research and testing.
“Given that these models learn best when they have lots of data, we hope our findings will lead to further collaborations with ICUs across Canada,” said Kazazian.
