ComEDA: A new tool for stress assessment based on electrodermal activity

Area of Science:

• Physiology
• Biomedical Engineering
• Data Science

Background:

• Sympathetic arousal responses to stress are complex and vary between physical and mental fatigue.
• Electrodermal activity (EDA) is a key non-invasive marker for sympathetic activation.
• Existing nonlinear analysis methods have limitations for EDA signal processing.

Purpose of the Study:

• To present ComEDA, a novel, parameter-insensitive approach for characterizing complex electrodermal activity dynamics.
• To overcome limitations of nonlinear analysis for ultra-short time series of EDA.
• To validate ComEDA's efficacy in distinguishing stress-induced physiological changes.

Main Methods:

• Developed ComEDA, a novel algorithm for complex dynamics characterization of EDA.
• Validated ComEDA using synthetic white and 1/f noise time series.
• Applied ComEDA to electrodermal activity signals from healthy subjects under physical and mental stress protocols.
• Compared ComEDA performance against Sample Entropy and other state-of-the-art metrics.

Main Results:

• ComEDA successfully discriminated increased complexity in 1/f noise compared to white noise (p < 0.03).
• ComEDA significantly differentiated between stressful tasks and resting states across all datasets (p < 0.01).
• The algorithm effectively distinguished varying complexity trends in EDA induced by physical versus mental stressors.
• ComEDA outperformed Sample Entropy in analyzing EDA signal complexity.

Conclusions:

• ComEDA is a robust and effective method for analyzing electrodermal activity complexity.
• The findings support ComEDA's potential for automated stress detection and monitoring of autonomic diseases.
• ComEDA offers a promising tool for telemedicine applications and understanding physiological responses to stress.