This research addresses exercise-related injuries by modeling individual physical capacity rather than relying on population averages. Using physiological and biomechanical data combined with machine learning, it aims to create personalized, dynamic thresholds for training. The goal is to prevent injury by aligning workload with real-time capacity, improving safety and long-term fitness outcomes.

This research introduces a sustainable, thread-based wearable device that measures lactate in sweat using chemiluminescence. By transforming cotton thread into a low-cost analytical tool, it enables simple, smartphone-based monitoring of physiological changes, offering an eco-friendly alternative to conventional biosensors for sports and health applications.

This research examined whether mental imagery training enhances neural efficiency and performance in basketball free throws. A single imagery session didn’t change EEG activity or performance overall, but higher confidence improved outcomes. Findings suggest imagery may boost performance indirectly through psychological factors, requiring longer or combined training for measurable neural effects.