Enhanced Respiratory Sinus Arrhythmia Quantification Using Variational Mode Decomposition and Multimodal Coupling Analysis for Emotion Recognition

Respiratory sinus arrhythmia (RSA) is a well-established physiological phenomenon that reflects autonomic nervous system (ANS) activity, possessing significant value in the realm of emotion recognition. Multimodal coupling analysis (MMCA) is a crucial algorithm for evaluating RSA, yet it faces limitations such as sensitivity to noise and non-stationary signals. To address these challenges, this study introduces the application of variational mode decomposition (VMD) to MMCA, leveraging VMD's ability to decompose complex signals into a series of band-limited intrinsic mode functions. The proposed VMD-MMCA approach aims to enhance the accuracy and robustness of RSA quantification, thereby improving emotion recognition capabilities. To validate the effectiveness of the proposed VMD-MMCA algorithm, a series of experiments were first conducted using simulation data. Subsequently, we applied the VMD-MMCA algorithm to investigate differences between anxious and calm states in individuals with spider phobia. Our results found that the proposed VMD-MMCA algorithm offers significant value in RSA quantification and emotion recognition. By improving the accuracy and robustness of RSA measurements, this method has the potential to advance our understanding of the physiological correlates of emotions and enhance the performance of emotion recognition systems.Clinical Relevance - This study introduces a novel VMD-MMCA algorithm for enhancing RSA quantification in emotion recognition, which may be of interest to practicing clinicians. By improving the accuracy of RSA measurements, this method could aid in the assessment and management of emotional disorders in clinical settings.