Neuropathic Pain Detection: An EEG-Based Brain Functional Network Approach.

Area of Science:

• Neuroscience
• Computational Biology
• Medical Diagnostics

Background:

• Chronic neuropathic pain (NP) diagnosis is challenging due to a lack of objective biomarkers, relying heavily on subjective evaluations.
• Existing diagnostic methods for NP are limited, hindering precise assessment and targeted treatment.
• Objective methods are needed to understand NP's underlying neural mechanisms and improve clinical outcomes.

Purpose of the Study:

• To develop an objective diagnostic approach for neuropathic pain using electroencephalogram (EEG) derived brain functional networks (BFNs).
• To investigate the utility of network control and topological features in identifying neural signatures of NP.
• To establish a machine learning-based classifier for distinguishing NP patients from healthy controls.

Main Methods:

• Analysis of EEG signals to compute BFNs using the weighted phase lag index (wPLI).
• Application of minimum dominating set (MDSet) analysis to assess network control properties.
• Extraction of persistent homology features and classification using a support vector machine (SVM).

Main Results:

• The developed SVM classifier achieved over 97% accuracy in binary classification between NP patients and control subjects.
• Specific neural connectivity patterns and network control disruptions associated with NP were identified.
• Classification accuracy decreased in distinguishing between different NP severity levels and in multiclass analysis.

Conclusions:

• EEG-based BFNs analysis offers a promising objective approach for diagnosing neuropathic pain.
• The findings reveal distinct neural signatures indicative of NP, advancing our understanding of the condition.
• This method has the potential to inform the development of more targeted therapeutic interventions for NP.