Key result
Machine learning applied to HRV identifies individual athletes with ~92% accuracy.
Why the study?
This study aimed to identify athletic characteristics using heart rate variability testing and machine learning algorithms.
Can machine learning algorithms applied to heart rate variability data accurately identify athletic characteristics and specific sports profiles?
Comparison
Athletes vs non-athletes (M1) and individual soccer player vs other team members (M2)
Design
Machine learning classification study
Authors
Loading...
Requires prospective validation before clinical or sports use; leaves open broader applicability of HRV-based ML models.
Observational
Can machine learning algorithms applied to heart rate variability data accurately identify athletic characteristics and specific sports profiles?
Effect estimate: Accuracy 0.84 (M1), 0.92 (M2)
Machine learning algorithms applied to heart rate variability data can accurately classify individuals as athletes or non-athletes and identify specific sports profiles.
Estrella et al. (2025) conducted an observational in Athleticism. Machine learning algorithms applied to heart rate variability was evaluated on Model performance (accuracy and ROC AUC) for classifying athletes vs non-athletes and identifying individual players (Accuracy 0.84 (M1), 0.92 (M2)). Machine learning applied to heart rate variability effectively classified athletes vs non-athletes (SVM accuracy 0.84, AUC 0.91) and identified individual soccer players (RF accuracy 0.92, AUC 0.94).