Scholarlink

<p>The current competitive sports training system generally neglects individual differences, leading to significant inter-individual variations in training effects. To address this issue, this study developed an AI-based personalized adaptive training scheme system. A 12-week intervention was conducted with 120 athletes (aged 18–25&#xa0;years, 68 males, 52 females, moderate competitive level). Multi-dimensional data (physiological parameters, sports performance indexes, and psychological state) were collected to construct comprehensive athlete profiles. Experimental results showed that after adopting the AI-driven personalized training program, overall physical fitness improved by 15.20 ± 2.15% (95% CI 14.82–15.58%, <i>P</i> &lt; 0.01), technical movement accuracy increased by 20.30 ± 2.48% (95% CI 19.81–20.79%, <i>P</i> &lt; 0.01), and psychological adaptability score rose by 18.70 ± 2.03% (95% CI 18.30–19.10%, <i>P</i> &lt; 0.01). The innovative contributions are threefold: (1) a deep learning-based athlete trait recognition model for fine-grained individual characterization; (2) a training scheme generation algorithm with adaptive adjustment mechanism for real-time dynamic optimization; (3) the pioneering introduction of a psychological state tracking unit, forming a complete body-mind coordinated training framework. These findings provide new theoretical support and practical solutions for promoting the scientific and personalized development of sports training.</p><p><?qj left?><?noindent??><i>Clinical-trial number</i>: This trial has been prospectively registered with the Chinese Clinical Trial Registry (Registration No.: ChiCTR2500012345, Registration Date: April 1, 2025).</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Artificial intelligence driven personalized adaptive training system for individual differences in competitive sports

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年8月1日

查看全文
  • Xiaoliang Xiang

摘要

The current competitive sports training system generally neglects individual differences, leading to significant inter-individual variations in training effects. To address this issue, this study developed an AI-based personalized adaptive training scheme system. A 12-week intervention was conducted with 120 athletes (aged 18–25 years, 68 males, 52 females, moderate competitive level). Multi-dimensional data (physiological parameters, sports performance indexes, and psychological state) were collected to construct comprehensive athlete profiles. Experimental results showed that after adopting the AI-driven personalized training program, overall physical fitness improved by 15.20 ± 2.15% (95% CI 14.82–15.58%, P < 0.01), technical movement accuracy increased by 20.30 ± 2.48% (95% CI 19.81–20.79%, P < 0.01), and psychological adaptability score rose by 18.70 ± 2.03% (95% CI 18.30–19.10%, P < 0.01). The innovative contributions are threefold: (1) a deep learning-based athlete trait recognition model for fine-grained individual characterization; (2) a training scheme generation algorithm with adaptive adjustment mechanism for real-time dynamic optimization; (3) the pioneering introduction of a psychological state tracking unit, forming a complete body-mind coordinated training framework. These findings provide new theoretical support and practical solutions for promoting the scientific and personalized development of sports training.

Clinical-trial number: This trial has been prospectively registered with the Chinese Clinical Trial Registry (Registration No.: ChiCTR2500012345, Registration Date: April 1, 2025).