Robustness of Symbolic Analysis for Estimating Heart-Rate Variability during Sport

Stress plays a crucial role in sports, influencing both physiological and psychological responses, specifically during competitions. Understanding the stress-response dynamics is critical for developing strategies to optimize performance and avoid adverse events, such as injury or death. Wearable sensors have revolutionized the real-time monitoring of athletes, being non-invasive and allowing the continuous collection of biomedical data during both training and competition. Heart rate variability (HRV), a key index of autonomic nervous system activity, has become an invaluable tool for assessing stress and recovery. Among various HRV analysis methods, symbolic analysis, which represents HRV data as discrete symbols based on signal amplitude or direction, seems to be robust in real-world conditions. The study aimed to evaluate the reliability and robustness of symbolic analysis in assessing stress induced by sports activities. Data was collected from 10 sprint athletes using portable sensors, recording 30-s electrocardiograms during various phases of training and competition. The symbolic analysis was applied to extract its symbolic patterns (0V, 1V, 2LV, 2UV) from RR-interval series at different times, that are rest, post-warm-up, end of short-distance running (eSDR), and recovery phases at 5-, 10-, and 15-minutes post-exercise, and from the same series by applying one-beat and two-beat delays. Statistical analysis revealed minimal significant differences between the different RR-interval series in terms of symbolic patterns, confirming the robustness of symbolic analysis to temporal variations. The strong correlation (ρ > 0.93) between the trends of symbolic patterns over time-related to the different RR-interval series supports the reliability of this technique in tracking autonomic regulation over time. These findings demonstrate the potential of symbolic analysis as a robust, real-time method for assessing stress in athletes, with practical implications for optimizing training and recovery strategies. Despite promising results, the study is limited in terms of sample size and variety of sports disciplines.