By: Huy Phi & Reza Hosseini Ghomi
Detecting Cognitive Changes Due to Concussion in Children Can be Challenging
Pediatric concussion is challenging to diagnose, particularly because children are constantly cognitively maturing and current diagnostic tools lack adequate sensitivity and specificity. Undetected concussion can be detrimental for a child’s brain development. This is important in the case of athletes, where athletes who express concussion symptoms must be removed from play to mitigate risk of further injury. Thus, an accessible and rapid cognitive test for the detection of concussion among children could be of great use. Our research team recently conducted a pilot study testing the utility of BrainCheck in detecting pediatric concussion, and have found promising results.
Pediatric Concussion and BrainCheck Study Design
We enrolled 126 healthy controls with no concussion history from Morton Ranch High School in Katy, TX, and 10 patients diagnosed with concussion from Texas Children’s Hospital. We had both groups take BrainCheck, where the concussion participants took BrainCheck within 96 hours of being admitted to the emergency department.
Results Demonstrate BrainCheck is Sensitive and Specific for Concussion in Children
Statistical analyses were performed to see if there were any significant differences in performance on BrainCheck among the concussion group compared to the control group. Our results show concussion patients had significantly diminished performance on BrainCheck Coordination, Immediate & Delayed Recognition, and Stroop tasks.
We then built a machine learning logistic regression model, which used a participant’s performance on BrainCheck to predict if they had a concussion or not. The model was able to achieve a sensitivity of 84% and specificity of 81% in detecting pediatric concussion. Despite the small sample size of the concussion group, BrainCheck was able to discriminate between concussed individuals from healthy with adequate accuracy.
At BrainCheck we are consistently working on numerous simultaneous studies to expand the applications of our tests and build the pool of data we use to determine if an individual is impaired. This study is one of many examples of our efforts to improve our product to help patients get the best care.
