Monitoring Training Load Responses in High-Level Swimmers: Heart Rate Variability, Sleep, Motivation, and Performance with Explainable Artificial Intelligence

Background: Understanding the impact of training sessions on physiological, psychological, and immunological responses is crucial for adequate training periodization and preventing negative influences on health, training, and performance. Heart rate variability has emerged as a valuable non-invasive marker of autonomic nervous system function and training adaptation, yet its relationship with other monitoring tools and performance outcomes during overload periods requires further investigation. The integration of artificial intelligence approaches may enhance the interpretation of complex training load data. Methods: This review synthesizes evidence from monitoring studies examining the responses of heart rate variability, sleep time and quality, motivation, dry-land strength, and swimming performance to overload periods in competitive swimmers. The capability of heart rate variability to assess daily variation in training loads is evaluated, with particular attention to the application of explainable artificial intelligence models. Evidence from studies utilizing orthostatic tests, Hooper index, sleep questionnaires, and rating of perceived exertion is systematically examined. Results: Evidence demonstrates that high-level swimmers accurately perceive their daily training loads; however, differences between prescribed training loads and perceived exertion loads emerge on weekends, indicating that physiological and psychological loads exert different influences and should be considered simultaneously when characterizing training loads. Overload periods characterized by increased training and perceived exertion loads do not necessarily elicit negative effects on sleep quantity and quality. Heart rate variability indices, particularly supine root mean square of successive differences and mean heart rate, emerge as the most sensitive markers of training load variation. Dry-land strength and swimming performance typically remain stable during overload periods, suggesting that the autonomic nervous system demonstrates greater sensitivity to acute and short-term training load changes than performance outcomes. Explainable artificial intelligence models demonstrate high predictive capability for assessing training load variations. Conclusion: Heart rate variability can be employed as a practical, sensitive tool for monitoring training responses and managing training loads in competitive swimmers. Integration of explainable artificial intelligence approaches enhances interpretation of complex monitoring data, supporting evidence-based training prescription and periodization decisions.

کلیدواژه ها:

Heart rate variability ، Swimming ، Training load ، Overload ، Artificial intelligence ، Athlete monitoring