Scholarlink

Background <p>Recovery is a vital multidimensional process for the athletic development of young football players, yet its assessment remains a significant methodological challenge. The rise of wearable technology and artificial intelligence presents revolutionary opportunities, although no systematic synthesis has assessed their integrated use within this specific developmental population.</p> Objective <p>Identify, catalog, and critically assess the scientific evidence regarding the use of technological tools and artificial intelligence applications for evaluating multidimensional recovery in football players aged 10–25&#xa0;years.</p> Methods <p>This systematic review was conducted in accordance with the PRISMA 2020 guidelines. A comprehensive bibliographic search was conducted in PubMed/MEDLINE, Web of Science, and Scopus (January 2015 to December 2025). The studies had to assess at least one dimension of recovery (neuromuscular, physiological, metabolic, psychological, sleep) using advanced technological tools and/or artificial intelligence algorithms in young football players. The methodological quality was assessed using the Newcastle–Ottawa Scale (observational studies) and the PEDro scale (randomized-controlled trials).</p> Results <p>Of the 139 identified records, 20 studies (with moderate-to-excellent quality) met the inclusion criteria. The most common technology was GPS/GNSS systems (68%), which were often combined with heart rate monitors and subjective questionnaires. Neuromuscular and psychophysiological recovery accounted for 70% of the areas assessed. Only 32% of the studies included machine learning algorithms, mainly Random Forest and XGBoost, which had classification accuracies of 79–91%. Multidimensional approaches have always been better than one-dimensional assessments.</p> Conclusions <p>Portable technologies and artificial intelligence present promising capabilities to revolutionize recovery monitoring among young footballers; however, significant methodological deficiencies remain regarding external validation, generalizability, and practical feasibility in resource-limited contexts.</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

Assessment of recovery in young soccer players: a systematic review of technological tools and artificial intelligence applications

  • 期刊论文

  • 发表日期：

    2026年7月1日

查看全文
  • Walid Bouzid,
  • Omar Ben Rakaa,
  • Younes Brouki,
  • Nada Ezzahra Rahnaoui,
  • Rochdi Daibi,
  • Aziz Chokri,
  • Aziz Eloirdi,
  • Carla Lourenço,
  • Mohamed Barkaoui

摘要

Background

Recovery is a vital multidimensional process for the athletic development of young football players, yet its assessment remains a significant methodological challenge. The rise of wearable technology and artificial intelligence presents revolutionary opportunities, although no systematic synthesis has assessed their integrated use within this specific developmental population.

Objective

Identify, catalog, and critically assess the scientific evidence regarding the use of technological tools and artificial intelligence applications for evaluating multidimensional recovery in football players aged 10–25 years.

Methods

This systematic review was conducted in accordance with the PRISMA 2020 guidelines. A comprehensive bibliographic search was conducted in PubMed/MEDLINE, Web of Science, and Scopus (January 2015 to December 2025). The studies had to assess at least one dimension of recovery (neuromuscular, physiological, metabolic, psychological, sleep) using advanced technological tools and/or artificial intelligence algorithms in young football players. The methodological quality was assessed using the Newcastle–Ottawa Scale (observational studies) and the PEDro scale (randomized-controlled trials).

Results

Of the 139 identified records, 20 studies (with moderate-to-excellent quality) met the inclusion criteria. The most common technology was GPS/GNSS systems (68%), which were often combined with heart rate monitors and subjective questionnaires. Neuromuscular and psychophysiological recovery accounted for 70% of the areas assessed. Only 32% of the studies included machine learning algorithms, mainly Random Forest and XGBoost, which had classification accuracies of 79–91%. Multidimensional approaches have always been better than one-dimensional assessments.

Conclusions

Portable technologies and artificial intelligence present promising capabilities to revolutionize recovery monitoring among young footballers; however, significant methodological deficiencies remain regarding external validation, generalizability, and practical feasibility in resource-limited contexts.