Intelligent optimization of track and field teaching using machine learning and wearable sensors

Area of Science:

• Sports Science
• Machine Learning in Education
• Biomechanics

Background:

• Traditional track and field education faces challenges with subjective assessments and manual feedback, hindering personalized instruction.
• Large-scale educational settings require innovative solutions for effective and individualized coaching.

Purpose of the Study:

• To develop and validate a novel machine learning framework for optimizing track and field teaching through intelligent analysis and personalization.
• To improve performance metrics and reduce injury risk in athletes via data-driven insights.

Main Methods:

• Developed an integrated multi-modal sensing system with wearable IMUs (200 Hz), high-definition cameras (120 fps), and force platforms.
• Implemented a hybrid CNN-BiLSTM architecture with ensemble learning for real-time biomechanical data analysis.
• Utilized reinforcement learning principles for an adaptive learning optimization algorithm.

Main Results:

• The hybrid CNN-BiLSTM ensemble model achieved F1-scores from 0.88 to 0.94 in sports classification tasks.
• The framework demonstrated a 27.3% improvement in time-to-proficiency and a 41.2% reduction in injury risk compared to traditional benchmarks.
• Ablation studies confirmed component synergies improved performance by 17.3% over individual subsystems.

Conclusions:

• The proposed machine learning framework offers a promising approach for data-driven pedagogical transformation in physical education.
• The system has potential for wide-scale implementation in postsecondary education and professional athletic training.
• This research establishes a foundation for intelligent, personalized coaching in sports.