Scholarlink

<p>The reciprocal benefits of physical exercise on physiological performance and mental well-being are well-documented, yet conventional training paradigms seldom optimize for both outcomes simultaneously. This results in a missed opportunity for synergistic adaptation, where physical gains bolster mental fortitude and vice versa. Here, we introduce and validate an AI-driven framework for generating personalized exercise regimens that explicitly co-optimize for athletic performance and psychological resilience. Our system is built on three core innovations: (1) a Dual-factor adaptation model (DFAM), a novel mathematical formalization of the coupled dynamics of physical fitness, fatigue, psychological stress, and resilience; (2) a personalized regimen optimization algorithm (PROA), a multi-objective evolutionary algorithm that navigates the trade-off space between performance and resilience to generate Pareto-optimal training schedules tailored to an individual’s unique DFAM parameters; and (3) an adaptive feedback control (AFC) system that fine-tunes daily training loads in real-time based on wearable sensor data and subjective feedback. We evaluated this framework in a 12-week randomized controlled trial with 60 competitive athletes. The personalized group (PER) demonstrated markedly superior gains in a composite Athletic Performance Index (API) (+ 21.4% ± 3.8%) compared to a standardized training group (STD) (+ 12.0% ± 4.5%) and a non-training control (CTL) (+ 2.2% ± 1.5%). Concurrently, psychological resilience, quantified via the Connor–Davidson resilience scale (CD-RISC), increased significantly more in the PER group (+ 18.6% ± 2.5%) than in the STD (+ 8.3% ± 2.1%) and CTL (+ 1.1% ± 1.3%) groups (all <i>p</i> &lt; 0.001). Ablation studies confirmed that both the dual-objective optimization and the adaptive feedback were critical for achieving these synergistic outcomes. Our findings suggest that an explicit, algorithmic co-optimization of physical and psychological states can significantly enhance holistic human performance, offering a promising new direction for intelligent training systems.</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

Synergistic enhancement of athletic performance and psychological resilience through AI-driven personalized exercise regimens network

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月14日

查看全文
  • Huayang Kang,
  • Yu Liu

摘要

The reciprocal benefits of physical exercise on physiological performance and mental well-being are well-documented, yet conventional training paradigms seldom optimize for both outcomes simultaneously. This results in a missed opportunity for synergistic adaptation, where physical gains bolster mental fortitude and vice versa. Here, we introduce and validate an AI-driven framework for generating personalized exercise regimens that explicitly co-optimize for athletic performance and psychological resilience. Our system is built on three core innovations: (1) a Dual-factor adaptation model (DFAM), a novel mathematical formalization of the coupled dynamics of physical fitness, fatigue, psychological stress, and resilience; (2) a personalized regimen optimization algorithm (PROA), a multi-objective evolutionary algorithm that navigates the trade-off space between performance and resilience to generate Pareto-optimal training schedules tailored to an individual’s unique DFAM parameters; and (3) an adaptive feedback control (AFC) system that fine-tunes daily training loads in real-time based on wearable sensor data and subjective feedback. We evaluated this framework in a 12-week randomized controlled trial with 60 competitive athletes. The personalized group (PER) demonstrated markedly superior gains in a composite Athletic Performance Index (API) (+ 21.4% ± 3.8%) compared to a standardized training group (STD) (+ 12.0% ± 4.5%) and a non-training control (CTL) (+ 2.2% ± 1.5%). Concurrently, psychological resilience, quantified via the Connor–Davidson resilience scale (CD-RISC), increased significantly more in the PER group (+ 18.6% ± 2.5%) than in the STD (+ 8.3% ± 2.1%) and CTL (+ 1.1% ± 1.3%) groups (all p < 0.001). Ablation studies confirmed that both the dual-objective optimization and the adaptive feedback were critical for achieving these synergistic outcomes. Our findings suggest that an explicit, algorithmic co-optimization of physical and psychological states can significantly enhance holistic human performance, offering a promising new direction for intelligent training systems.