Concussions are the most common type of brain injury in pediatric patients and can result in persistent post-concussion symptoms and alter developmental capabilities. To make matters worse, objectively assessing traumatic brain injuries in concussed patients is a challenge. In an attempt to find a tool of diagnostic utility, a team of Minneapolis researchers discovered that an automated eye-tracking algorithm serving as a biomarker for concussions correlates with symptoms, convergence insufficiency and accommodative dysfunction associated with concussions in the pediatric population.

This cross-sectional, case-control study evaluated 56 concussed children and 83 healthy controls. Researchers obtained metrics comparing velocity and conjugacy of eye movements over time and compared them with the correlation between Acute Concussion Evaluation (ACE) scores, convergence and accommodation.

The team found that 12 eye-tracking metrics were significantly different between concussed and non-concussed children. They note that a model to classify concussions as diagnosed by symptoms using the ACE had an area-under-the-curve (AUC) of 0.854. The researchers add that an eye-tracking model built to identify near-point-of-convergence (NPC) disability achieved 95.8% specificity and 57.1% sensitivity for an AUC of 0.810. They also include that reduced binocular amplitude of accommodation had a Spearman correlation of 0.752 with NPC.