Real-Time High-Level Acute Pain Detection Using a Smartphone and a Wrist-Worn Electrodermal Activity Sensor

Area of Science:

• Biomedical Engineering
• Physiological Measurement
• Digital Health

Background:

• Subjective pain assessment leads to medication misuse, addiction, and overdose.
• Real-time, objective pain quantification is needed, especially for at-home monitoring.
• Ambulatory devices are crucial for continuous pain assessment outside clinical settings.

Purpose of the Study:

• To develop and evaluate a smartphone application for objective, real-time pain detection.
• To implement electrodermal activity (EDA) indices for pain quantification.
• To assess the accuracy of algorithms in distinguishing between pain and painless states.

Main Methods:

• Utilized a wrist-worn device to collect electrodermal activity (EDA) signals.
• Implemented time- and frequency-domain EDA indices within a smartphone application.
• Validated algorithms using thermal grill (n=10) and electrical pulse (n=15) pain stimuli, calibrated to visual analog scale (VAS) levels of 8/10 and 7/10, respectively.

Main Results:

• All evaluated EDA features significantly differed between pain and painless segments.
• Random forest algorithm achieved 81.5% accuracy in pain detection.
• High sensitivity (78.9%) and specificity (84.2%) were observed using leave-one-subject-out cross-validation.

Conclusions:

• Smartphone-based EDA analysis offers a viable method for objective pain detection.
• The developed system demonstrates potential for near real-time, ambulatory pain monitoring.
• This technology could aid in more accurate pain management and reduce opioid-related risks.