Clinical EEG Recordings Research Articles

Alpha coherence is a network signature of cognitive recovery from disorders of consciousness.

Alpha (8-12 Hz) frequency band oscillations are among the most informative features in electroencephalographic (EEG) assessment of patients with disorders of consciousness (DoC). Because interareal alpha synchrony is thought to facilitate long-range communication in healthy brains, coherence measures of resting-state alpha oscillations may provide insights into a patients capacity for higher-order cognition beyond channel-wise estimates of alpha power. In multi-channel EEG, global coherence methods may be used to augment standard spectral analysis methods by both estimating the strength and identifying the structure of coherent oscillatory networks. We performed global coherence analysis in 95 separate clinical EEG recordings (28 healthy controls and 33 patients with acute or chronic DoC, 25 of whom returned for follow-up) collected between two academic medical centers. We found that posterior alpha coherence is associated with recovery of higher-level cognition. We developed a measure of network organization, based on the distance between eigenvectors of the alpha cross-spectral matrix, that detects recovery of posterior alpha networks. In patients who have emerged from a minimally conscious state, we showed that coherence-based alpha networks are reconfigured prior to restoration of alpha power to resemble those seen in healthy controls. This alpha network measure performs well in classifying recovery from DoC (AUC = 0.78) compared to common representations of functional connectivity using the weighted phase lag index (AUC = 0.50 - 0.57). Lastly, we observed that activity within these alpha networks is suppressed during positive responses to task-based EEG command-following paradigms, supporting the potential utility of this biomarker to detect covert cognition. Our findings suggest that restored alpha networks may represent a sensitive early signature of cognitive recovery in patients with DoC. Therefore, network detection methods may augment the utility of EEG assessments for DoC.

Adaptive threshold algorithm for detecting EEG-interburst intervals in extremely preterm neonates.

Objective. This study provides an adaptive threshold algorithm for burst detection in electroencephalograms (EEG) of preterm infantes and evaluates its performance using clinical real-world EEG data.Approach. We developed an adaptive threshold algorithm for burst detection in EEG recordings from preterm infants. To assess its applicability in the real-world, we tested the algorithm on a dataset of 30 clinical EEG recordings which were not preselected for good quality, to ensure a real-world scenario.Main results. Interrater agreement was substantial at a kappa of 0.73 (0.68-0.79 inter-quantile range). The performance of the algorithm showed a similar agreement with one clinical expert of 0.73 (0.67-0.76) and a sensitivity and specificity of 0.90 (0.82-0.94) and 0.95 (0.93-0.97), respectively.Significance. The adaptive threshold algorithm demonstrated robust performance in detecting burst patterns in clinical EEG data from preterm infants, highlighting its practical utility. The fine-tuned algorithm achieved similar performance to human raters. The algorithm proves to be a valuable tool for automated burst detection in the EEG of preterm infants.

CNN-Based Pattern Classifiers for Precise Identification of Perinatal EEG Biomarkers of Brain Injury in Preterm Neonates

Electroencephalographic (EEG) monitoring is important for the diagnosis of hypoxic-ischemic (HI) brain injury in high-risk preterm infants. EEG monitoring is limited by the reliance on expert clinical observation. However, high-risk preterm infants often do not present observable symptoms due to their frailty. Thus, there is an urgent need to find better ways to automatically quantify changes in the EEG these high-risk babies. This article is a first step towards this goal. This innovative study demonstrates the effectiveness of deep Convolutional Neural Networks (CNN) pattern classifiers, trained on spectrally-detailed Wavelet Scalograms (WS) images derived from neonatal EEG sharp waves—a potential translational HI biomarker, at birth. The WS-CNN classifiers exhibit outstanding performance in identifying HI sharp waves within an exclusive clinical EEG recordings dataset of preterm infants immediately after birth. The work has impact as it demonstrates exceptional high accuracy of 99.34 ± 0.51% cross-validated across 13,624 EEG patterns over 48 h raw EEG at low 256 Hz clinical sampling rates. Furthermore, the WS-CNN pattern classifier is able to accurately identify the sharp-waves within the most critical first hours of birth (n = 8, 4:36 ± 1:09 h), regardless of potential morphological changes influenced by different treatments/drugs or the evolutionary ‘timing effects’ of the injury. This underscores its reliability as a tool for the identification and quantification of clinical EEG sharp-wave biomarkers at bedside.

Etiological and risk factors in recurrent febrile seizures: Insights through EEG analysis.

Febrile seizures, convulsive episodes in young children during febrile illnesses, are a significant concern due to their potential for recurrence and associated uncertainties. This study investigated the causes and risks associated with recurrent febrile seizures and the critical role of electroencephalogram (EEG) in their accurate diagnosis. Following Institutional Review Board approval and going through the consenting process with parents, this study gathered the clinical features and EEG recordings of children admitted with febrile seizures in the Department of Pediatrics, Mahatma Gandhi Memorial Hospital, Kakatiya Medical College, Warangal, Telangana, India. Descriptive statistics, including mean, standard deviation (SD), frequencies, and percentages, were computed to understand the data comprehensively. The Chi-Square test was employed to analyze the association between variables, with a significance level of 0.05, ensuring reliable and trustworthy findings. Out of 42 children studied, 28 (66.67%) presented with simple febrile seizures, with the mean time of occurrence of seizures from the onset of fever being 7.85 hours. Abnormal EEG was seen in 50% of children with complex febrile seizures and 16% with simple febrile seizures. Generalized epileptiform discharges were the most common epileptic activity observed. Low sodium levels had a significant relationship with febrile seizures in the analysis. This study emphasizes the importance of EEG in diagnosing febrile seizures, particularly in complex cases. Our findings suggest that low sodium levels may be a significant risk factor for febrile seizures. Further research is necessary to identify other preventable risk factors to reduce the burden of the medical condition.

Entraining brain oscillatory activity at alpha frequencies in dementia with Lewy bodies

AbstractBackgroundDementia with Lewy bodies (DLB) is characterized by a marked shift of electroencephalographic (EEG) power and dominant rhythm, from the alpha towards the theta frequency range. Transcranial alternate current stimulation (tACS) is a novel non‐invasive brain stimulation technique that allows entrainment of cerebral oscillations at desired frequencies. The objective of the present study was to evaluate safety and efficacy of occipital alpha tACS in DLB patients.MethodWe performed a double‐blind, randomized, sham‐controlled, cross‐over clinical trial in 12 DLB patients. Patients were randomized to receive real (60 minutes of 3 mA peak‐to‐peak stimulation at 12 Hz) or sham stimulation over the occipital cortex (Oz according to the 10‐20 international EEG system). Patients underwent clinical evaluation (clock‐drawing test, trail‐making test part A and B, the quantitative pentagon drawing test subitem), neurophysiological assessment (short‐latency afferent inhibition to evaluate cholinergic circuits) and EEG recordings, at baseline and after real/sham stimulation.ResultOccipital alpha tACS was safe and well tolerated in all patients. We observed a significant increase in clinical performance scores, in neurophysiological measures of cholinergic transmission and in occipital alpha power spectrum density after real stimulation but not sham stimulation.ConclusionOccipital tACS delivered at alpha frequencies may be a novel therapeutic approach in patients with DLB.

A Two-Stage Automatic System for Detection of Interictal Epileptiform Discharges from Scalp Electroencephalograms.

The objective of this work was to develop a deep learning-based automatic system with reliable performance in detecting interictal epileptiform discharges (IEDs) from scalp electroencephalograms (EEGs). For the present study, 484 raw scalp EEG recordings were included, standardized, and split into 406 for training and 78 for testing. Two neurophysiologists individually annotated the recordings for training in channel-wise manner. Annotations were divided into segments, on which nine deep neural networks (DNNs) were trained for the multiclassification of IED, artifact, and background. The fitted IED detectors were then evaluated on 78 EEG recordings with IED events fully annotated by three experts independently (majority agreement). A two montage-based decision mechanism (TMDM) was designed to determine whether an IED event occurred at a single time instant. Area under the precision-recall curve (AUPRC), as well as false-positive rates, F1 scores, and kappa agreement scores for sensitivity = 0.8 were estimated. In multitype classification, five DNNs provided one-versus-rest AUPRC mean value >0.993 using fivefold cross-validation. In IED detection, the system that had integrated the temporal convolutional network (TCN)-based IED detector and the TMDM rule achieved an AUPRC of 0.811. The false positive was 0.194/min (11.64/h), and the F1 score was 0.745. The agreement score between the system and the experts was 0.905. The proposed framework provides a TCN-based IED detector and a novel two montage-based determining mechanism that combined to make an automatic IED detection system. The system would be useful in aiding clinic EEG interpretation.

The potential of the Lempel-Ziv complexity of the EEG in diagnosing cognitive impairment in patients with temporal lobe epilepsy.

To analyze whether the Lempel-Ziv Complexity (LZC) in quantitative electroencephalogram differs between the temporal lobe epilepsy (TLE) patients with or without cognitive impairment (CI) and explore the diagnostic value of LZC for identifying CI in TLE patients. Twenty-two clinical features and 20-min EEG recordings were collected from 48 TLE patients with CI and 27 cognitively normal (CON) TLE patients. Seventy-six LZC features were calculated for 19 leads in four frequency bands (alpha, beta, delta, and theta). The clinical and LZC features were compared between the two groups. A support vector machine (SVM) was subsequently constructed using the leave-one-out method of cross-validation for LZC features with statistical differences. Regarding the clinical features, the level of education (p < .001), hippocampal atrophy and sclerosis (p = .029), and depression (p = .037) were statistically different between the two groups. For the LZC features, there were statistically significant differences in the alpha (Fp1, Fz, Cz, Pz, C3, C4, T3, T4, T5, T6, F3, F4, F7, F8, O1, and O2), beta (Fp2), and theta (F7) oscillations. The mean LZC in the alpha band was higher in the TLE-CI group than that in the CON group, and there were no differences in the remaining bands. The SVM model showed 74.51% accuracy, 79.63% sensitivity, 84.30% F1 score, 68.75% specificity, and .85 area under the curve scores. The LZC in the alpha band might have the potential to be used as a biomarker for the diagnosis of TLE combined with CI. The TLE-CI group, on the other hand, exhibited a higher degree of complexity in alpha oscillations, which were widespread and occurred in all brain regions.

Prevalence, electroclinical spectrum and effect on the outcome of non-convulsive status epilepticus in critically ill patients; the utility of routine electroencephalogram

ObjectivesTo highlight the importance of routine electroencephalogram (rEEG) in detecting non-convulsive status epilepticus (NCSE), describing the electroclinical spectrum and effect on outcome in critically ill patients with altered mental status (CIPAMS). MethodsThis retrospective study was conducted at King Fahd University Hospital. Clinical data and EEG recordings of CIPAMS to rule out NCSE were reviewed. All patients had at least 30 minutes of EEG recording. The Salzburg Consensus criteria (SCC) were applied to diagnose NCSE. The data analysis was performed using SPSS version 22.0. The chi-squared test was used to compare categorical variables such as etiologies, EEG findings, and functional outcomes. Multivariable analysis was performed to identify the predictors of unfavorable outcomes. ResultsA total of 323 CIPAMS referred to rule out NCSE were enrolled (mean age 57.8 ± 20 years). Nonconvulsive status epilepticus was diagnosed in 54 (16.7%) patients. A significant association was found between subtle clinical features and NCSE (P =< 0.01). Acute ischemic stroke (18.5%), sepsis (18.5%), and hypoxic brain injury (22.2%) were the main etiologies. The previous history of epilepsy was significantly associated with NCSE (P = 0.01). Acute stroke, cardiac arrest, mechanical ventilation, and NCSE were statistically associated with unfavorable outcomes. Nonconvulsive status epilepticus was an independent predictor of unfavorable outcomes (P = 0.02, OR = 2.75, CI = 1.16–6.48) during multivariable analysis. Sepsis was associated with higher mortality (P =< 0.01, OR = 2.4, CI = 1.4–4.0). SignificanceOur study findings suggest that the utility of rEEG in detecting NCSE in CIPAMS should not be underestimated. Important observations further indicate that repeating rEEG is advisable, as this would increase the likelihood of identifying NCSE. Thus, physicians should consider and repeat rEEG when evaluating CIPAMS in order to detect NCSE, which is an independent predictor of unfavorable outcomes. Nonetheless, further studies comparing rEEG and cEEG yields are required to augment the current understanding of the electroclinical spectrum and better describe NCSE in CIPAMS.

Reflex seizures induced by micturition: a case report

BackgroundReflex seizures (RS) induced by micturition are extremely rare, and the clinical and electroencephalogram features of RS are not widely known among clinicians. In particular, the origin of the epileptic area is still unclear.Case presentationAn 8-year-old girl who had generalized tonic-clonic seizures was diagnosed with RS induced by micturition based on the clinical manifestation and EEG recordings. We also reviewed the clinical and EEG characteristics of RS induced by micturition in literature by searching the databases of PubMed and MEDLINE using keywords “micturition reflex seizure”, “reflex seizure induced by micturition”, and “micturition induced seizure” by January 2022. We speculate that the mechanism of micturition-induced RS may involve excessive neuronal excitation in regions that participate in micturition.ConclusionsThe RS in this patient was considered to be induced by micturition. Awareness should be raised to this rare form of RS among practitioners.

Time-Frequency Complexity Maps for EEG-Based Diagnosis of Alzheimer’s Disease Using a Lightweight Deep Neural Network

Alzheimer’s disease (AD) is a neurodegenerative disorder with an unknown etiology and a significant prevalence. Rapid and accurate detection of AD is crucial to assist in a more effective and tailored treatment plan to delay the progression of the disease. This paper introduces a novel approach based on a time-frequency complexity map (complextrogram) for the automated AD diagnosis. The complextrogram is the topographic complexity level of an EEG signal, plotted as a function of time and frequency. The complextrogram representations were fed into a well-known lightweight deep neural network called MobileNet for robust performance on resource and accuracy tradeoffs. The experiments were performed using a five-fold cross-validation technique on a publicly available database containing clinical EEG recordings from 24 patients with AD and 24 healthy, age-matched controls. The proposed pipeline provided competitive performance with just 2.2 M parameters and achieved the best overall accuracy for some locations in the frontal lobes (Fp2 and F8 channels). For both channels, the classification accuracy was 100%. Also, the violin plot was used to get further details of the distribution of complexity values for specific frequency rhythms. After statistical evaluation, it was observed that neurodegenerative conditions caused changes in chaotic behaviors, including increased delta complexity and decreased alpha complexity. Results demonstrated that the complextrogram representation proved its potency for the input quality required by the deep learning architectures. Furthermore, the complextrogram method is a promising pathway to discriminate and reflect the fundamental characteristics of AD abnormalities.

Developing Disposable EEG Cap for Infant Recordings at the Neonatal Intensive Care Unit.

Long-term EEG monitoring in neonatal intensive care units (NICU) is challenged with finding solutions for setting up and maintaining a sufficient recording quality with limited technical experience. The current study evaluates different solutions for the skin–electrode interface and develops a disposable EEG cap for newborn infants. Several alternative materials for the skin–electrode interface were compared to the conventional gel and paste: conductive textiles (textured and woven), conductive Velcro, sponge, super absorbent hydrogel (SAH), and hydro fiber sheets (HF). The comparisons included the assessment of dehydration and recordings of signal quality (skin interphase impedance and powerline (50 Hz) noise) for selected materials. The test recordings were performed using snap electrodes integrated into a forearm sleeve or a forehead band along with skin–electrode interfaces to mimic an EEG cap with the aim of long-term biosignal recording on unprepared skin. In the hydration test, conductive textiles and Velcro performed poorly. While the SAH and HF remained sufficiently hydrated for over 24 h in an incubator-mimicking environment, the sponge material was dehydrated during the first 12 h. Additionally, the SAH was found to have a fragile structure and was electrically prone to artifacts after 12 h. In the electrical impedance and recording comparisons of muscle activity, the results for thick-layer HF were comparable to the conventional gel on unprepared skin. Moreover, the mechanical instability measured by 1–2 Hz and 1–20 Hz normalized relative power spectrum density was comparable with clinical EEG recordings using subdermal electrodes. The results together suggest that thick-layer HF at the skin–electrode interface is an effective candidate for a preparation-free, long-term recording, with many advantages, such as long-lasting recording quality, easy use, and compatibility with sensitive infant skin contact.

Digital reconstruction of infraslow activity in human intracranial ictal recordings using a deconvolution-based inverse filter

Infraslow activity (ISA) is a biomarker that has recently become of interest in the characterization of seizure recordings. Recent data from a small number of studies have suggested that the epileptogenic zone may be identified by the presence of ISA. Investigation of low frequency activity in clinical seizure recordings, however, has been hampered by technical limitations. EEG systems necessarily include a high-pass filter early in the measurement chain to remove large artifactual drifts that can saturate recording elements such as the amplifier. This filter, unfortunately, attenuates legitimately seizure-related low frequencies, making ISA difficult to study in clinical EEG recordings. In this study, we present a deconvolution-based digital inverse filter that allows recovery of attenuated low frequency activity in intracranial recordings of temporal lobe epilepsy patients. First, we show that the unit impulse response (UIR) of an EEG system can be characterized by differentiation of the system’s step response. As proof of method, we present several examples that show that the low frequency component of a high-pass filtered signal can be restored by deconvolution with the UIR. We then demonstrate that this method can be applied to biologically relevant signals including clinical EEG recordings obtained from seizure patients. Finally, we discuss how this method can be applied to study ISA to identify and assess the seizure onset zone.

Neuropathology findings in KCNQ2 neonatal epileptic encephalopathy.

PurposeKCNQ2-epileptic encephalopathy (EE) is a neonatal epilepsy syndrome characterized by a typical clinical presentation and EEG recording, but without any brain or cortical abnormal development on MRI. Most of the patients have a severe developmental impairment. The epileptogenic mechanisms are thought to be the result of the changes of the M-current density causing a change of brain excitability. Although recent studies allow us to better understand the physiopathology of KCNQ2-EE, the neuropathology of this ion channel dysfunction has only been previously described in a single case report. MethodsWe report the neuropathology study of a case of KCNQ2-EE with a typical electro-phenotype due to a de novo heterozygous single nucleotide pathogenic variant in the exon 5 of the KCNQ2 gene (NM_172107.2:c.802C>T; p.Leu268Phe). ResultsAt the macroscopic level, the brain had a normal structure with a normal neocortical gyral pattern. At the histological level, the cortex had a usual six-layer lamination in all lobes but blurred gray-white matter boundaries due to excessive heterotopic neurons in deep white matter were observed. This diffuse mild malformation of cortical development is suggestive of a neuronal migration disorder. ConclusionIn recent years, our understanding of the role of ion channel dysfunctions in early brain development has expanded from the occurrence of EE to brain malformation. Through this rare neuropathological report, we emphasize the role of KCNQ2 channels in the process of cortical development. As for other genetic neonatal onset epilepsies, more reports are needed to further delineate the range of neuropathological abnormalities for KCNQ2-EE.

VE-CAM-S: Visual EEG-Based Grading of Delirium Severity and Associations With Clinical Outcomes.

This prospective, single-center, observational cohort study was conducted from August 2015 to December 2016 and October 2018 to December 2019. Academic medical center, all inpatient wards. Adult inpatients undergoing a clinical EEG recording; excluded if deaf, severely aphasic, developmentally delayed, non-English speaking (if noncomatose), or if goals of care focused primarily on comfort measures. Four-hundred six subjects were assessed; two were excluded due to technical EEG difficulties. None. A machine learning model, with visually coded EEG features as inputs, was developed to produce scores that correlate with behavioral assessments of delirium severity (Confusion Assessment Method-Severity [CAM-S] Long Form [LF] scores) or coma; evaluated using Spearman R correlation; area under the receiver operating characteristic curve (AUC); and calibration curves. Associations of Visual EEG Confusion Assessment Method Severity (VE-CAM-S) were measured for three outcomes: functional status at discharge (via Glasgow Outcome Score [GOS]), inhospital mortality, and 3-month mortality. Four-hundred four subjects were analyzed (mean [sd] age, 59.8 yr [17.6 yr]; 232 [57%] male; 320 [79%] White; 339 [84%] non-Hispanic); 132 (33%) without delirium or coma, 143 (35%) with delirium, and 129 (32%) with coma. VE-CAM-S scores correlated strongly with CAM-S scores (Spearman correlation 0.67 [0.62-0.73]; p < 0.001) and showed excellent discrimination between levels of delirium (CAM-S LF = 0 vs ≥ 4, AUC 0.85 [0.78-0.92], calibration slope of 1.04 [0.87-1.19] for CAM-S LF ≤ 4 vs ≥ 5). VE-CAM-S scores were strongly associated with important clinical outcomes including inhospital mortality (AUC 0.79 [0.72-0.84]), 3-month mortality (AUC 0.78 [0.71-0.83]), and GOS at discharge (0.76 [0.69-0.82]). VE-CAM-S is a physiologic grading scale for the severity of symptoms in the setting of delirium and coma, based on visually assessed electroencephalography features. VE-CAM-S scores are strongly associated with clinical outcomes.

EEG normal variants: A prospective study using the SCORE system

ObjectiveTo determine the prevalence and characteristics of normal variants in EEG recordings in a large cohort, and provide readers with typical examples of all normal variants for educational purposes. MethodsUsing the SCORE EEG system (Standardized Computer-Based Organized Reporting of EEG), we prospectively extracted EEG features in consecutive patients. In this dataset, we analyzed 3050 recordings from 2319 patients (mean age 38.5 years; range: 1–89 years). ResultsThe distribution of the normal variants was as follows: sharp transients 19.21% (including wicket spikes), rhythmic temporal theta of drowsiness 6.03%, temporal slowing of the old 2.89%, slow fused transients 2.59%, 14-and 6-Hz bursts 1.83%, breach rhythm 1.25%, small sharp spikes 1.05%, 6-Hz spike and slow wave 0.69% and SREDA 0.03%. ConclusionsThe most prevalent normal variants are the sharp transients, which must not be over-read as epileptiform discharges. SignificanceEEG readers must be familiar with the normal variants to avoid misdiagnosis and misclassification of patients referred to clinical EEG recordings.

Topography-Related EEG-fMRI in Surgically Confirmed Epileptic Foci: A Comparison to Spike-Related EEG-fMRI in Clinical Practice.

EEG-fMRI has gained increasing importance in epilepsy pre-surgical diagnosis. However, 40-70% of EEG-fMRI recordings in patients lack interictal epileptiform discharges (IEDs) during the scan, which could be overcome by detecting matching topography maps. We tried to validate this method in clinical settings taking various electroclinical factors into consideration. Eleven patients who had undergone EEG-fMRI during pre-surgical evaluation for drug-resistant epilepsy and who had had clinical long-term video-EEG were studied. Spike-related blood oxygen level-dependent (BOLD) maps were created using IEDs occurring during the EEG-fMRI scan. Separate maps were then generated from IEDs marked on the clinical long-term EEG recordings, which were averaged to produce topographical IED maps and correlated with the EEGs recorded inside the scanner yielding a correlation coefficient time course. Epileptogenic zones were defined by an expert panel during pre-surgical evaluation and validated by an epilepsy surgery resulting in a good outcome. Both techniques' performance was evaluated according to factors including arousal during IED recording, IED topography and lateralization, lesion type, and localization. Topography-related EEG-fMRI yielded more specific results compared to the spike-related method. Superficial lesion location and ipsilateral IED seem to result in a higher concordance of BOLD maps. The polarity of BOLD responses may be lesion-dependent, and both positive and negative BOLD changes may be associated with the irritative zone. Topography-related EEG-fMRI may show improved specificity especially for superficial lesions producing ipsilateral spikes. This method can be used as an alternative either in the absence of spikes during the simultaneous EEG-fMRI acquisition or to sharpen a diffusely activated BOLD-map.

Resting-State Electroencephalography for Prognosis in Disorders of Consciousness Following Traumatic Brain Injury.

Although the majority of patients recover consciousness after a traumatic brain injury (TBI), a minority develop a prolonged disorder of consciousness, which may never fully resolve. For these patients, accurate prognostication is essential to treatment decisions and long-term care planning. In this review, we evaluate the use of resting-state electroencephalography (EEG) as a prognostic measure in disorders of consciousness following TBI. We highlight that routine clinical EEG recordings have prognostic utility in the short to medium term. In particular, measures of alpha power and variability are indicative of relatively better functional outcomes within the first year post-TBI. This is hypothesized to reflect intact thalamocortical loops, and thus the potential for recovery of consciousness even in the apparent absence of current consciousness. However, there is a lack of research into the use of resting-state EEG for predicting longer-term recovery following TBI. We conclude that, given the potential for patients to demonstrate improvements in consciousness and functional capacity even years after TBI, a research focus on EEG-augmented prognostication in very long-term disorders of consciousness is now required.

EEG Microstate Differences in Medicated vs. Medication-Naïve First-Episode Psychosis Patients

There has been considerable interest in the role of synchronous brain activity abnormalities in the pathophysiology of psychotic disorders and their relevance for treatment; one index of such activity are EEG resting-state microstates. These reflect electric field configurations of the brain that persist over 60–120 ms time periods. A set of quasi-stable microstates classes A, B, C, and D have been repeatedly identified across healthy participants. Changes in microstate parameters coverage, duration and occurrence have been found in medication-naïve as well as medicated patients with psychotic disorders compared to healthy controls. However, to date, only two studies have directly compared antipsychotic medication effects on EEG microstates either pre- vs. post-treatment or between medicated and unmedicated chronic schizophrenia patients. The aim of this study was therefore to directly compare EEG resting-state microstates between medicated and medication-naïve (untreated) first-episode (FEP) psychosis patients (mFEP vs. uFEP). We used 19-channel clinical EEG recordings to compare temporal parameters of four prototypical microstate classes (A–D) within an overall sample of 47 patients (mFEP n = 17; uFEP n = 30). The results demonstrated significant decreases of microstate class A and significant increases of microstate class B in mFEP compared to uFEP. No significant differences between groups were found for microstate classes C and D. Further studies are needed to replicate these results in longitudinal designs that assess antipsychotic medication effects on neural networks at the onset of the disorder and over time during illness progression. As treatment response and compliance in FEP patients are relatively low, such studies could contribute to better understand treatment outcomes and ultimately improve treatment strategies.

EEG microstates as biomarker for psychosis in ultra-high-risk patients

Resting-state EEG microstates are brief (50–100 ms) periods, in which the spatial configuration of scalp global field power remains quasi-stable before rapidly shifting to another configuration. Changes in microstate parameters have been described in patients with psychotic disorders. These changes have also been observed in individuals with a clinical or genetic high risk, suggesting potential usefulness of EEG microstates as a biomarker for psychotic disorders. The present study aimed to investigate the potential of EEG microstates as biomarkers for psychotic disorders and future transition to psychosis in patients at ultra-high-risk (UHR). We used 19-channel clinical EEG recordings and orthogonal contrasts to compare temporal parameters of four normative microstate classes (A–D) between patients with first-episode psychosis (FEP; n = 29), UHR patients with (UHR-T; n = 20) and without (UHR-NT; n = 34) later transition to psychosis, and healthy controls (HC; n = 25). Microstate A was increased in patients (FEP & UHR-T & UHR-NT) compared to HC, suggesting an unspecific state biomarker of general psychopathology. Microstate B displayed a decrease in FEP compared to both UHR patient groups, and thus may represent a state biomarker specific to psychotic illness progression. Microstate D was significantly decreased in UHR-T compared to UHR-NT, suggesting its potential as a selective biomarker of future transition in UHR patients.

Machine Learning with Imbalanced EEG Datasets using Outlier-based Sampling.

Epilepsy is a neurological disorder which causes seizures in over 65 million people worldwide. Recently developed implantable therapeutic devices aim to prevent symptoms by applying acute electrical stimulation to the seizure-generating brain region in response to activity detected by on-device machine learning hardware. Many training algorithms require an equal number of examples for each target class (e.g. normal activity and seizures), and performance can suffer if this condition is not satisfied. In the case of epilepsy, poor performance can cause seizures to be missed, or stimulation to be applied erroneously. As there is an abundance of normal (interictal) data in clinical EEG recordings, but seizures are rare events (less than 1% of the dataset), the data available for training is severely imbalanced. There are several conventional pre-processing methods used to address imbalanced class learning, such as down-sampling of the majority class and up-sampling of the minority class, but each have performance drawbacks. This paper presents an improved method which involves reducing the majority class down to the most effective interictal outlier samples. Outliers are determined by using Exponentially Decaying Memory Signal Energy (EDMSE) features with Isolation Forests and an ANOVA-based method, which involves comparing a moving feature window to a baseline reference window. Outlier-based sampling is tested with two classifiers (KNN and Logistic Regression) and achieves higher accuracy (∼2% increase) and fewer false positives (∼38% decrease), along with a lower latency (∼3 seconds shorter) compared to conventional training set pre-processing methods.