Common Spatial Pattern EEG decomposition for Phantom Limb Pain detection

Abstract

Phantom Limb Pain (PLP) is a chronic condition frequent among individuals with acquired amputation. PLP has been often investigated with the use of functional MRI focusing on the changes that take place in the sensorimotor cortex after amputation. In the present study, we investigated whether a different type of data, namely electroencephalographic (EEG) recordings, can be used to study the condition. We acquired resting state EEG data from people with and without PLP and then used machine learning for a binary classification task that differentiates the two. Common Spatial Pattern (CSP) decomposition was used as the feature extraction method and two validation schemes were followed for the classification task. Six classifiers (LDA, Log, QDA, LinearSVC, SVC and RF) were optimized through grid search and their performance compared. Two validation approaches, namely all-subjects validation and leave-one-out cross-validation (LOOCV), resulted in high classification accuracy. Most notably, the 93.7% accuracy achieved with SVC in LOOCV holds promise for good diagnostic capabilities using EEG biomarkers. In conclusion, our findings indicate that EEG data is a promising target for future research aiming at elucidating the neural mechanisms underlying PLP and its diagnosis.

Translational Applications of Neuroimaging

“…The following types of biomarkers have the potential to be clinically applicable in chronic pain management (Tracey et al, 2019;Van Der Miesen et al, 2019) A combination of these biomarkers is also a possible outcome for future research. Meneses et al, 2016;Lancaster et al, 2017;Ferdek et al, 2019) Decreased alpha band power (2/8; diagnostic) (Vachon-Presseau et al, 2016;Telkes et al, 2020) Increased theta band power (2/5; monitoring) (Graversen et al, 2012; Alpha band activity (1/4; prognostic) (Vuckovic et al, 2018) Frontal delta power (1/3; predictive) (Yüksel et al, 2019) Decreased peak alpha frequency (2/3; diagnostic, monitoring) (Baliki et al, 2012;Uygur-Kucukseymen et al, 2020) Decreased alpha band power application of SVM classifiers has led to an improvement in accuracy with up to 93.7% Kragel et al, 2018;Buchanan et al, 2021;Lendaro et al, 2021;Zolezzi et al, 2021;Teel et al, 2022;Topaz et al, 2022). With gradual improvements in the ML algorithms over the years, there is a trend of testing their applicability in clinical practice, especially for diagnostic, monitoring, and prognostic purposes in the context of chronic pain (Mendonça-de-Souza et al, 2012;Sufianov et al, 2014).…”

“…The main concern arises from the complex, dynamic nature of pain which limits the ability to capture EEG signals of prognostic value, requiring EEG recordings over a longer period. Nonetheless, there is a possibility that features learned through ML in studies Increased alpha band power (1/1; diagnostic) (Cao et al, 2018) Higher prefrontal complexity in the preictal phase (1/2; prognostic) (Thibaut et al, 2017) Increased frontal and central gamma band power (1/2; diagnostic) (Wei et al, 2022) Increased theta band power (1/2; diagnostic) (Lendaro et al, 2021) Increased prefrontal gamma band power (1/2; diagnostic, monitoring) (Barbosa-Torres and Cubo-Delgado, 2021) using other neuroimaging modalities, such as fMRI (Baliki et al, 2012), can be used to monitor the transition between disease states from continuous EEG signals to be used as monitoring biomarkers. Meanwhile, only a few studies have examined this possibility in a longitudinal cohort of pain patients.…”

Application of EEG-based Machine Learning in Stock Trading-related Emotion Recognition

“…Its fundamental principle involves the projection of data into lower dimensions. In this projection, the goal is to make the projected points of each data category as close as possible while maximizing the separation distance [20] between the category centers of different data categories. The primary objective of LDA is to minimize the ratio of intraclass variance and, concurrently, maximize the inter-class discrimination.…”

In search of a composite biomarker for chronic pain by way of EEG and machine learning: where do we currently stand?

“…Moreover, the decoder performed only at 57% accuracy—close to chance-level, leaving much room for improvement. In more recent years, researchers’ primary goal in improving decoding performance has been motivated primarily by the goal of optimizing model generalization, where the application of SVM classifiers has led to an improvement in accuracy with up to 93.7% ( Misra et al, 2017 ; Kragel et al, 2018 ; Levitt et al, 2020 ; Buchanan et al, 2021 ; Lendaro et al, 2021 ; Zolezzi et al, 2021 ; Teel et al, 2022 ; Topaz et al, 2022 ).…”

“…The following types of biomarkers have the potential to be clinically applicable in chronic pain management (Tracey et al, 2019;Van Der Miesen et al, 2019) A combination of these biomarkers is also a possible outcome for future research. Decreased alpha band power (5/18; diagnostic, predictive) (Sufianov et al, 2014;González-Roldán et al, 2016;Meneses et al, 2016;Lancaster et al, 2017;Ferdek et al, 2019) Decreased alpha band power (2/8; diagnostic) (Vachon-Presseau et al, 2016;Telkes et al, 2020) Increased theta band power (2/5; monitoring) (Graversen et al, 2012;Jensen et al, 2013) Alpha band activity (1/4; prognostic) (Vuckovic et al, 2018) Frontal delta power (1/3; predictive) (Yüksel et al, 2019) Decreased peak alpha Decreased beta band power (1/18; diagnostic, predictive) (Lancaster et al, 2017) Increased theta global network efficiency (1/8; monitoring) (Teixeira et al, 2021) Decreased SMR/theta power ratio with neuro-feedback therapy (1/18; monitoring) (Di Pietro et al, 2018) (Continued) Rockholt et al 10.3389/fnins.2023.1186418 Frontiers in Neuroscience 12 frontiersin.org application of SVM classifiers has led to an improvement in accuracy with up to 93.7% (Misra et al, 2017;Kragel et al, 2018;Levitt et al, 2020;Buchanan et al, 2021;Lendaro et al, 2021;Zolezzi et al, 2021;Teel et al, 2022;Topaz et al, 2022). With gradual improvements in the ML algorithms over the years, there is a trend of testing their applicability in clinical practice, especially for diagnostic, monitoring, and prognostic purposes in the context of chronic pain (Mendonça-de-Souza et al, 2012;Sufianov et al, 2014).…”