The Vanderbilt Lab for Immersive AI Translation (VALIANT) is collaborating on research that is using machine learning and artificial intelligence to more accurately determine if a person has a mild traumatic brain injury (TBI), or concussion.
The research is being funded by a $1.4 million U.S. Department of Defense grant titled, “Diagnosis of Mild TBI Spectrum via Assessment Battery and Machine Learning.”
Researchers say determining whether a TBI is mild can be challenging because of a lack of findings. For instance, a person who suffers a head-related injury may report having a problem with vision, but the CT scan or MRI is negative. As a result, they’re diagnosed as having a mild TBI.
Tonia Rex is principal investigator on the grant and professor of ophthalmology and visual sciences at Vanderbilt University School of Medicine. She believes that a battery of assessments is needed for the best diagnosis, and that applying machine learning and AI approaches can help pinpoint the problem.
“When we were looking at brain regions and structural differences, we only saw changes in three areas in three people in our previous study,” says Rex, holder of the Marlene and Spencer Hays Directorship in Translational Vision Research at the Vanderbilt Eye Institute. “When we applied machine learning, we could detect 70% of the TBI individuals. So, we’re talking 11% verses 70%; a huge improvement in detecting differences.”
VALIANT is led by Bennett Landman, a preeminent scholar who holds the Stevenson Chair of Electrical and Computer Engineering and has joint appointments in computer science, biomedical engineering, radiology and radiological sciences, psychiatry and behavioral sciences, biomedical informatics, and neurology.
Landman, who is a Vanderbilt lead investigator, says the project is an example of the radical collaboration that has become a mission of Vanderbilt University.
“Mild traumatic brain injury is a critical national health problem in both civilian and military areas,” says Landman. “Our project will make real progress toward personalizing understanding and treatment of individual symptoms using new areas of AI.”
Other collaborators include Adam Anderson, professor of biomedical engineering; Martin Gallagher, assistant professor of neurology, VUMC; Reid Longmuir, assistant professor of ophthalmology and visual sciences, VUMC; and Lucas Groves, ophthalmologist, Blanchfield Hospital, Fort Campbell.
Contact: Lucas Johnson, lucas.l.johnson@vanderbilt.edu