A Comprehensive Machine-Learning-Based Software Pipeline to Classify EEG Signals: A Case Study on PNES vs. Control Subjects

Abstract

The diagnosis of psychogenic nonepileptic seizures (PNES) by means of electroencephalography (EEG) is not a trivial task during clinical practice for neurologists. No clear PNES electrophysiological biomarker has yet been found, and the only tool available for diagnosis is video EEG monitoring with recording of a typical episode and clinical history of the subject. In this paper, a data-driven machine learning (ML) pipeline for classifying EEG segments (i.e., epochs) of PNES and healthy controls (CNT) is introduced. This software pipeline consists of a semiautomatic signal processing technique and a supervised ML classifier to aid clinical discriminative diagnosis of PNES by means of an EEG time series. In our ML pipeline, statistical features like the mean, standard deviation, kurtosis, and skewness are extracted in a power spectral density (PSD) map split up in five conventional EEG rhythms (delta, theta, alpha, beta, and the whole band, i.e., 1–32 Hz). Then, the feature vector is fed into three different supervised ML algorithms, namely, the support vector machine (SVM), linear discriminant analysis (LDA), and Bayesian network (BN), to perform EEG segment classification tasks for CNT vs. PNES. The performance of the pipeline algorithm was evaluated on a dataset of 20 EEG signals (10 PNES and 10 CNT) that was recorded in eyes-closed resting condition at the Regional Epilepsy Centre, Great Metropolitan Hospital of Reggio Calabria, University of Catanzaro, Italy. The experimental results showed that PNES vs. CNT discrimination tasks performed via the ML algorithm and validated with random split (RS) achieved an average accuracy of 0.97 ± 0.013 (RS-SVM), 0.99 ± 0.02 (RS-LDA), and 0.82 ± 0.109 (RS-BN). Meanwhile, with leave-one-out (LOO) validation, an average accuracy of 0.98 ± 0.0233 (LOO-SVM), 0.98 ± 0.124 (LOO-LDA), and 0.81 ± 0.109 (LOO-BN) was achieved. Our findings showed that BN was outperformed by SVM and LDA. The promising results of the proposed software pipeline suggest that it may be a valuable tool to support existing clinical diagnosis.

Resting-state EEG for the diagnosis of idiopathic epilepsy and psychogenic nonepileptic seizures: A systematic review

“…Diagnostic accuracy was explored by two separate studies. Descriptive indices of power achieved high discriminative performance in one study (acc: 0.81-0.99; [49]), and Lagged Functional Connectivity was reported to be a good predictor of diagnosis by the second study (sens: 0.67, spec: 0.67; [50]). As no validation was performed on novel samples, no information on generalizability of these models is available.…”

“…In four studies (based on two fully independent study samples), a comparable number of males and females were examined. In two studies, the patient sample had higher prevalence of females [23,49]. All participants were not taking any medications in three studies [23,48,49], while in the sample shared by the remaining three studies most patients were taking AEDs, benzodiazepines or antidepressant medications [46,47,50].…”

“…All studies used a case-control design. Four studies had group-level descriptors as their only outcome [23,46,47,48], one focused on diagnostic accuracy indices only [49], and one examined both [50].…”

Real-Time Artifacts Reduction during TMS-EEG Co-Registration: A Comprehensive Review on Technologies and Procedures

“…In this setting, complex feedback, through open or closed loops, can give further insights into the functional mechanics of cortical activity in the human brain, and has much experimental and therapeutic potential [ 176 , 177 ]. Capturing higher quality TMS-EEG data, with a reduction in noise, has become a slightly less arduous task, and has allowed researchers to develop novel techniques to identify and understand patterns of clinical significance [ 178 , 179 , 180 , 181 , 182 , 183 , 184 , 185 ]. We believe our survey highlights challenges and proposes solutions related to TMS-EEG experiments.…”

Signatures of Brain Network Alteration in Psychogenic Non-Epileptic Seizures: A Rest-EEG Study Based on Power Spectral Density and Phase Lag Index

“…In terms of spontaneous scalp EEGs, a general trend is to investigate the redistribution of power in the frequency bands to grasp the peak of frequency or dominant band. Fourier-based power spectral density (PSD), to characterize the frequency distribution of resting-state PNES via EEG data, was used only in [4]. Functional connectivity within particular resting-state networks can be used to study local organization and global integrated and spatially distributed architecture.…”