Functional Asymmetries in Youth Soccer Players: A Systematic Review of Assessment Technologies and Artificial Intelligence Applications

Background. Functional asymmetries in youth soccer players represent an important biomechanical and neuromuscular phenomenon, acting both as adaptive mechanisms and as potential risk factors for injury. However, existing evidence remains fragmented regarding the effectiveness of technological and artificial intelligence (AI)–based approaches for their assessment across different populations. Purpose. This systematic review aimed to synthesize current methodologies and applications of technological tools and artificial intelligence used to assess functional asymmetries in youth soccer players. Methods. The review was conducted in accordance with PRISMA guidelines. A total of 418 records were identified through searches in Scopus, Web of Science, and PubMed, of which 28 studies met the inclusion criteria. Methodological quality was evaluated using standardized assessment tools. The included studies examined wearable technologies, biomechanical assessment systems, and AI-based analytical models for asymmetry detection and interpretation. Results. AI-based models demonstrated moderate predictive accuracy for injury risk, with reported values up to AUC = 0.70. Wearable inertial measurement units showed high ecological validity for on-field assessment of functional asymmetries. However, substantial heterogeneity was observed across studies in terms of asymmetry thresholds, outcome definitions, and analytical procedures, limiting cross-study comparability. A key finding was that strength asymmetries exceeding 15% were significantly associated with increased injury risk (p < 0.05), while such asymmetries did not consistently impair sprint performance, indicating context-dependent functional effects. Conclusions. The findings indicate that technological and AI-based approaches provide valuable tools for assessing functional asymmetries in youth soccer players, particularly when applied in ecologically valid settings. The integration of explainable AI with multimodal biomechanical data appears to be a promising direction for personalized monitoring and intervention strategies. Future research should prioritize standardized methodological reporting, harmonized asymmetry thresholds, and longitudinal study designs to improve comparability and clinical translatability of findings.