Classifying athletes and non-athletes by differences in spontaneous brain activity: a machine learning and fMRI study

Area of Science:

• Neuroscience
• Sports Science
• Brain Imaging

Background:

• Sports training induces brain changes, but commonalities across disciplines and training duration effects are unclear.
• Understanding these neural adaptations is crucial for optimizing athletic performance and cognitive function.

Purpose of the Study:

• To investigate commonalities in brain activity changes across different sports disciplines.
• To explore the relationship between brain activity alterations and the duration of sports training.
• To differentiate athletes from non-athletes based on brain activity patterns.

Main Methods:

• Resting-state functional magnetic resonance imaging (rs-fMRI) was used to analyze spontaneous brain activity.
• Amplitude of low-frequency fluctuations (ALFF) and fractional amplitude of low-frequency fluctuations (fALFF) were employed as key metrics.
• Machine learning algorithms were utilized for classification of brain activity patterns.

Main Results:

• Athletes showed significantly higher ALFF in the Right Insula, Right Posterior orbital gyrus, and Right Lateral orbital gyrus compared to controls.
• Controls exhibited higher fALFF in the Right Postcentral gyrus.
• A negative correlation was found between fALFF in the Right Postcentral gyrus and years of professional training in athletes.
• Machine learning accurately classified athletes from non-athletes with over 96.97% accuracy.

Conclusions:

• Functional brain reorganization in athletes may represent an adaptation to prolonged training, potentially enhancing processing efficiency.
• Long-term sports training impacts brain function, influencing cognitive and sensory systems vital for athletic performance.
• Machine learning shows promise for identifying athletes based on distinct brain activity profiles.