Using Symmetric Projection Attractor Reconstruction to Visualise and Analyse Prosodic Action-based ECG Morphologies...

Music playing engages the autonomic nervous system and affects the cardiovascular system. Understanding the interaction between playing and cardiovascular response requires measuring cardiovascular signals in ways that do not limit musicians' actions. Noise due to movements and nonstandard electrode placements may also constrain the analysis of ECG signal morphologies, requiring methods that eliminate the need for precise fiducial point annotation. One promising approach is the Symmetric Projection Attractor Reconstruction (SPAR) method, which converts periodic time series into a two-dimensional map called an attractor. This study applies the SPAR method to musicians' ECG signals, enabling visualisation and classification of signal morphologies. A professional trio-violinist, cellist, and pianist-rehearses Schubert's Trio No. 2, Op. 100, Andante con moto nine times. Their ECG signals are segmented based on the musicians' annotated prosodic structures. The results show clear differences between attractors during prosodic structures and those during the baseline. Hierarchical clustering of attractor density profiles first separates the data by musician, then by performance state (baseline vs. music), and finally by the type of musical action. When comparing performances over time, a significant negative correlation is found between baseline and music states in string players. This suggests a reduction in attractor distance, indicating physiological adaptation over time. Our study introduces SPAR as a robust way to analyse ECG morphology during music performance and link attractor patterns with prosodic action types.Clinical relevance-Music-based interventions present valuable potential as safe, adjunctive cardiovascular treatment to conventional therapies. Music engagement, both active and receptive, modulates the autonomic nervous system. Robust methods for representing and understanding the physiological dynamics of music making is an important step towards applying music-based interventions in real-world settings.