Scholarlink

<p>Youth basketball talent development in China faces multiple challenges—homogeneous training methods, insufficient attention to individual differences, and a lack of systematic, evidence-based decision-making. This study protocol describes the design and planned validation of an AI- and big-data-driven personalized training system aimed at testing whether integrated multidimensional data and dynamic feedback can improve talent identification and training outcomes in Chinese youth basketball. The protocol outlines the randomized effectiveness trial, process evaluation, and implementation measures that will be used to evaluate feasibility, potential efficacy, and adoption. A mixed-methods, randomized controlled pretest–posttest design will be implemented. One hundred youth basketball athletes aged 12–18 will be recruited and randomized (1:1) to an AI-driven personalized training (experimental) arm or a conventional training (control) arm. Multidimensional data (physical, technical, training-load, psychological, and game statistics) will be collected using validated field tests, wearable devices, video analysis, and standardized questionnaires. Primary quantitative analyses will compare pre–post changes between groups using intention-to-treat principles. The primary inferential approach will use ANCOVA (adjusting for baseline score, age stratum, sex, and team clustering) and linear mixed-effects models for repeated measures; effect sizes (Cohen’s d) and 95% CIs will be reported. Advanced methods (Bayesian regression, SEM) and predictive ML models will be treated as exploratory pending sample size adequacy and diagnostic checks. Qualitative interviews with coaches and participants will explore acceptability and implementation barriers guided by the Technology Acceptance Model (TAM). This study is anticipated to meet the criteria for exemption under Article 32 of the 2023 Ethical and Moral Management Measures; nevertheless, the full protocol and participant-facing materials will be submitted to the institutional Ethics Review Committee for formal confirmation (or exemption letter) and any correspondence will be archived. Study findings will be disseminated for academic, non-commercial purposes subject to ethical approvals.</p><p><i>Trial registration</i> To be registered prior to trial commencement. Registry name and registration number will be provided in the final manuscript (e.g., Chinese Clinical Trial Registry or ClinicalTrials.gov).</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

AI and big data personalized training protocol for Chinese youth basketball

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月9日

查看全文
  • Anqi Wu,
  • Aowei Zhang,
  • Chao Zhou

摘要

Youth basketball talent development in China faces multiple challenges—homogeneous training methods, insufficient attention to individual differences, and a lack of systematic, evidence-based decision-making. This study protocol describes the design and planned validation of an AI- and big-data-driven personalized training system aimed at testing whether integrated multidimensional data and dynamic feedback can improve talent identification and training outcomes in Chinese youth basketball. The protocol outlines the randomized effectiveness trial, process evaluation, and implementation measures that will be used to evaluate feasibility, potential efficacy, and adoption. A mixed-methods, randomized controlled pretest–posttest design will be implemented. One hundred youth basketball athletes aged 12–18 will be recruited and randomized (1:1) to an AI-driven personalized training (experimental) arm or a conventional training (control) arm. Multidimensional data (physical, technical, training-load, psychological, and game statistics) will be collected using validated field tests, wearable devices, video analysis, and standardized questionnaires. Primary quantitative analyses will compare pre–post changes between groups using intention-to-treat principles. The primary inferential approach will use ANCOVA (adjusting for baseline score, age stratum, sex, and team clustering) and linear mixed-effects models for repeated measures; effect sizes (Cohen’s d) and 95% CIs will be reported. Advanced methods (Bayesian regression, SEM) and predictive ML models will be treated as exploratory pending sample size adequacy and diagnostic checks. Qualitative interviews with coaches and participants will explore acceptability and implementation barriers guided by the Technology Acceptance Model (TAM). This study is anticipated to meet the criteria for exemption under Article 32 of the 2023 Ethical and Moral Management Measures; nevertheless, the full protocol and participant-facing materials will be submitted to the institutional Ethics Review Committee for formal confirmation (or exemption letter) and any correspondence will be archived. Study findings will be disseminated for academic, non-commercial purposes subject to ethical approvals.

Trial registration To be registered prior to trial commencement. Registry name and registration number will be provided in the final manuscript (e.g., Chinese Clinical Trial Registry or ClinicalTrials.gov).