Big Data-Driven Personalized Training for Mental Health Enhancement in Sports
Bushan Wan, Xuejun Ma, Qingshan DuanAbstract
Objective
This study aims to develop and evaluate a big data-driven personalized training system for athletes, with a specific focus on assessing its impact on psychological well-being. The research investigates how data-optimized training prescriptions influence not only physical performance but also key mental health indicators, including stress, mood, and emotional resilience.
Subjects and Methods
A cohort of 80 competitive athletes participated in a 16-week intervention. A multi-source data platform continuously collected training load, physiological responses, sleep, and self-reported mood/stress data. Machine learning algorithms generated dynamically adjusted, personalized training plans. Mental health was assessed at baseline, midpoint, and post-intervention using the Profile of Mood States (POMS), the Perceived Stress Scale (PSS), and heart rate variability (HRV) monitoring.
Results
Athletes following the personalized training regimen showed a 28% greater improvement in target performance metrics (p<0.01) compared to a matched control group on standard periodized plans. Concurrently, they demonstrated a 35% reduction in PSS scores (p<0.001) and a 40% improvement in positive mood scores. HRV data indicated significantly better autonomic nervous system recovery. The algorithm successfully identified individual "overtraining risk" patterns linked to declining mood states, enabling pre-emptive adjustments.
Conclusions
Big data-driven personalization in sports training effectively enhances both athletic performance and psychological well-being. By synchronizing training loads with an athlete's psychophysiological state, this approach fosters a more sustainable and mentally healthy athletic development model, reducing burnout risk and promoting holistic athlete health.
Corresponding Author
Xuejun Ma, Qufu Normal University, Qufu 273165, China.