Structural Magnetic Resonance Imaging Brain Age Investigation in Athletes with Persistent Postconcussion Syndrome.

Area of Science:

• Neuroimaging
• Artificial Intelligence
• Sports Medicine

Background:

• Brain age prediction algorithms estimate biological brain age from structural MRI, identifying deviations from normal aging.
• Traumatic brain injury and sports-related concussions are linked to an increased brain age gap (BAG).
• Persistent postconcussion syndrome (PCS) may be associated with altered brain aging trajectories.

Purpose of the Study:

• To investigate the brain age gap (BAG) in athletes with persistent postconcussion syndrome (PCS+) compared to those without (PCS-).
• To utilize SHapley Additive exPlanations (SHAP) to identify features driving brain age predictions.
• To explore the relationship between cumulative sports-related concussions (SRCs) and brain age.

Main Methods:

• Brain age was calculated from T1-weighted MRI scans in 50 athletes (24 with PCS+), aged 22-73.
• SHapley Additive exPlanations (SHAP), an explainable AI framework, was employed.
• Clinical variables and the number of self-reported SRCs were collected.

Main Results:

• Athletes with PCS+ exhibited a brain age approximately 5 years older than PCS- athletes.
• No clinical variable was significantly associated with the observed brain age gap.
• A greater brain age was noted in athletes reporting five or more SRCs.
• SHAP analysis identified the third ventricle as a key feature for PCS+ and the superior temporal sulcus posterior for PCS-.

Conclusions:

• Brain age prediction and explainable AI show promise in studying PCS in athletes.
• PCS+ is associated with accelerated brain aging, potentially linked to cumulative head trauma.
• Further research is needed to understand the mechanisms and identify biomarkers for early intervention.