EEG-Based Tonic Cold Pain Characterization Using Wavelet Higher Order Spectral Features.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Tonic cold pain characterization relies on subjective measures.
• Existing electroencephalograph (EEG) analysis methods often lack sensitivity to nonlinear brain responses.
• Phase-blind spectral analysis limits the capture of complex neural dynamics during pain.

Purpose of the Study:

• To introduce and validate a novel EEG analysis approach using wavelet higher order spectral (WHOS) features for tonic cold pain characterization.
• To extend current pain assessment capabilities by dynamically monitoring nonlinearities in brain responses.
• To objectively quantify pain levels and transitions using EEG data.

Main Methods:

• Analysis of EEG data from 17 healthy volunteers subjected to tonic cold pain stimuli.
• Application of wavelet higher order spectral (WHOS) analysis to capture quadratic phase coupling.
• Utilizing bifrequency wavelet bispectrum/bicoherence for nonlinear dynamics monitoring.
• Classification analysis using four established classifiers to assess feature discrimination.

Main Results:

• WHOS-based features effectively discriminated between relaxation and pain states.
• The method successfully identified transitions from relaxation to mild and severe pain.
• Subjective pain perception was translated into objective measures of pain endurance.
• Demonstrated the ability to capture changes in nonlinear brain activity correlating with pain intensity.

Conclusions:

• The proposed WHOS-based EEG analysis offers a promising, objective method for tonic cold pain characterization.
• This approach advances beyond relative power spectrum analysis by incorporating phase information.
• Findings suggest potential applications for WHOS in assessing other pain types, including chronic pain and various clinical scenarios.