Severity identification of chronic neuropathic pain based on EEG analysis
Montemayor Zolezzi, Daniela
MetadataShow full item record
The lack of an integral characterization of chronic neuropathic pain (NP) has led to pharmacotherapy mismanagement and has hindered advances in clinical trials. In this study, we attempted to identify chronic NP by integrating psychometric (based on Brief Inventory of Pain – BIP), and both linear and nonlinear electroencephalographic (EEG) features. For this purpose, 35 chronic NP patients were firstly recruited voluntarily. All of the volunteers filled in the BIP; and additionally, 22 EEG channels positioned in accordance with the 10/20 international system were registered for 10 minutes at resting state: 5 minutes with eyes open and 5 minutes with eyes closed. EEG Signals were sampled at 250Hz within a bandwidth between 0.1 and 100Hz. Secondly, linear and non-linear EEG features were extracted from healthy and NP conditions. In addition, a database of an age matched control group of 13 healthy participants was downloaded from the Figbase open access site. As linear features, band power was obtained per clinical band considering five regions: prefrontal, frontal, central, parietal and occipital. As non-linear features, approximate entropy was calculated per channel and per clinical frequency band. The control group was used as an EEG feature reference against those in NP condition to explain power and entropy tendencies. Finally, resulting feature vectors in NP condition were grouped in line with the BIP outcome to create BIP-EEG patterns in three groups: (a) low, (b) moderate, and (c) high pain. Resulting BIP-EEG patterns were classified achieving 96% accuracy for all severities, F-score of 95% for high pain, and 94% F-score for moderate and low pain. Taken together, these results indicate that EEG activity of chronic NP patients shows significant differentiable neuroplastic trends with respect to the severity of pain in the spontaneous state for each clinical band. The main contribution of this work is proving ApEn as a method that effectively characterizes the different levels of chronic NP. With further characterization of NP, entropy might be an appropriate and sufficient method to monitor the experience of pain and aid physicians to achieve a better pain management and treatment for chronic NP.