Development of Machine Learning Algorithms Using EEG Data to Detect the Presence of Chronic Pain

Chronic pain impacts more than one in five adults in the United States (US) and the costs associated with the condition amount to hundreds of billions of dollars annually. Despite the tremendous impact of chronic pain globally, the standard of care for diagnosis depends on subjective self-reporting of pain state, with no objective assessment procedure available. This study investigated the application of signal processing and machine learning to electroencephalography (EEG) data for the development of classification algorithms capable of differentiating subjects with diverse chronic pain etiologies from pain-free subjects. The study population included participants experiencing various types of chronic pain, including nociceptive, neuropathic, and mixed etiological pain conditions. Chronic pain diagnoses were based on clinical evaluation by participating physicians and adhered to the International Association for the Study of Pain (IASP) definition, requiring pain persistence for >3 months and associated functional impairment or emotional distress. Data from 186 participants were used for algorithm development, including 35 healthy controls and 151 chronic pain patients. Machine learning methodologies were applied to the data, with Elastic Net chosen as the optimal methodology.. The classifier was able to differentiate pain versus no pain subjects with an accuracy of 79.6%, sensitivity of 82.2%, and specificity of 66.7%. This study incorporates the multidimensional nature of chronic pain, ensuring that our methods and interpretations align with current clinical and research standards. This study represents a step toward integrating EEG-based biomarkers into clinical workflows for chronic pain assessment, bridging the gap between subjective reporting and objective diagnostic tools.