Differentiating between stress- and EPT-induced electrodermal activity during dental examination

Area of Science:

• Neuroscience
• Biomedical Engineering
• Pain Management

Background:

• Dental pain activates the sympathetic nervous system, measurable via electrodermal activity (EDA).
• Accurate pain quantification in dentistry is crucial for optimizing treatments and reducing drug addiction risks.
• Residual stress can confound EDA measurements during pain assessment.

Purpose of the Study:

• To investigate if EDA can differentiate between stress and pain during dental examinations.
• To utilize electrical pulp testing (EPT) as a method to simulate pain with residual stress.
• To develop a machine learning approach for discriminating stress from EPT-induced pain signals.

Main Methods:

• Fifty-one subjects underwent EPT on vital and necrotic teeth.
• Four data groups were defined: mild stress, mild stress + EPT, strong stress, and strong stress + EPT.
• A multilayer perceptron model was used to analyze EDA-derived features.

Main Results:

• EDA features showed significant differences between stress + EPT groups and stress-only groups.
• The machine learning model achieved 84.6% accuracy, 76.2% sensitivity, and 86.8% specificity.
• EPT stimulation resulted in greater EDA amplitude and faster responses compared to stress alone.

Conclusions:

• Machine learning algorithms can effectively discriminate between stress and EPT stimulation using EDA signals.
• This approach holds potential for objective pain assessment in clinical dental settings.
• Differentiating stress from pain via EDA could lead to more precise pain management strategies.