The electroencephalography (EEG) signal is a noninvasive and complex signal that has numerous applications in biomedical fields, including sleep and the brain-computer interface. Given its complexity, researchers have proposed several advanced preprocessing and feature extraction methods to analyze EEG signals. In this study, we analyze a comprehensive review of numerous articles related to EEG signal processing. We searched the major scientific and engineering databases and summarized the results of our findings. Our survey encompassed the entire process of EEG signal processing, from acquisition and pretreatment (denoising) to feature extraction, classification, and application. We present a detailed discussion and comparison of various methods and techniques used for EEG signal processing. Additionally, we identify the current limitations of these techniques and analyze their future development trends. We conclude by offering some suggestions for future research in the field of EEG signal processing.
Decades of research have shown machine learning superiority in discovering highly nonlinear patterns embedded in electroencephalography (EEG) records compared with conventional statistical techniques. However, even the most advanced machine learning techniques require relatively large, labeled EEG repositories. EEG data collection and labeling are costly. Moreover, combining available datasets to achieve a large data volume is usually infeasible due to inconsistent experimental paradigms across trials. Self-supervised learning (SSL) solves these challenges because it enables learning from EEG records across trials with variable experimental paradigms, even when the trials explore different phenomena. It aggregates multiple EEG repositories to increase accuracy, reduce bias, and mitigate overfitting in machine learning training. In addition, SSL could be employed in situations where there is limited labeled training data, and manual labeling is costly. This article: 1) provides a brief introduction to SSL; 2) describes some SSL techniques employed in recent studies, including EEG; 3) proposes current and potential SSL techniques for future investigations in EEG studies; 4) discusses the cons and pros of different SSL techniques; and 5) proposes holistic implementation tips and potential future directions for EEG SSL practices.
BACKGROUND: Conventional electroencephalography remains the gold standard for the diagnosis and quantification of neonatal seizures. However, amplitude-integrated electroencephalography (aEEG) is being introduced to neonatal intensive care as an adjunct for neonatal seizure detection. OBJECTIVES: This study's purpose was to determine the sensitivity of neonatal seizure detection in a single electroencephalogram channel (C3-->C4), used to simulate the raw signal from which aEEG is derived. We also aimed to determine the sensitivity of seizure detection by neonatologists by using aEEG and to establish those neonatal seizure characteristics that are associated with their correct detection by aEEG. METHODS: Conventional electroencephalograms with neonatal seizures were reviewed for electroencephalogram background and neonatal seizure characteristics (site of onset, duration, and peak-to-peak amplitude). The presence, duration, and peak-to-peak amplitude of each seizure were simultaneously noted in a single electroencephalogram channel (C3-->C4). aEEGs generated from this channel were reviewed for background and seizures by 6 neonatologists with varying aEEG interpretation expertise. RESULTS: A total of 851 neonatal seizures from 125 conventional electroencephalograms were analyzed. The patients' conceptional ages were 34 to 50 weeks. Because 94% of the conventional electroencephalograms had > or = 1 neonatal seizure visible in C3-->C4, and 78% of all neonatal seizures appeared in the C3-->C4 channel, the theoretical sensitivity of seizure detection in a single electroencephalogram channel was high. However, seizures were briefer and lower in amplitude in C3-->C4 compared with conventional electroencephalography. Neonatologists identified seizures in 22% to 57% of the 125 records of neonatal seizure. They detected 12% to 38% of the 851 individual seizures. Multivariate analysis revealed that the appearance of seizures in C3-->C4, neonatal seizure duration, seizure amplitude, seizure count per hour, and neonatologists' experience with aEEG interpretation all correlated with neonatal seizure detection. CONCLUSIONS: Even among physicians who have extensive experience, many neonatal seizures are difficult to detect on an aEEG, especially when they are infrequent, brief, or of low amplitude.
Current Practice of Clinical Electroencephalography, 4e - Libros de Medicina - Neurologia general - 145,00
This book deals with the field of Electroencephalography in the widest possible sense, from the cellular foundations of the electric activities of the brain to a vast number of clinical applications. The basic science sections were up-dated to include advanced computer modeling approaches. The chapters on normal and pathological EEG findings in premature infants, newborns and children were thoroughly revised to keep up with the advances that have taken place recently in studying brain developmental issues. Major advances have taken place in neurophysiological findings in a variety of neurodegenerative disorders, which led to thoroughly revised chapters. Other rapidly changing subjects related to EEG recording/monitoring in ICU's, EMUs, and operating rooms, in patients with epilepsy, head injuries, infectious disorders and those undergoing surgical procedures, led to radically updating a number of chapters and to the addition of a chapter dedicated to invasive recordings for the treatment of patients with movement disorders. A previously missing chapter on the neurophysiology of myoclonus was added. Chapters that deal with automated EEG interpretation techniques and with standardizing EEG reporting using ILAE/IFCN approved terminology, were also added. Many chapters in the on-line version of the book will have the ability to link to a database of over 150 complete EEGs that cover the scope seen in a general EEG Lab. This link will allow the reader to manipulate the EEG display parameters as if they were in their own lab, generate a report and compare it to one generated by a panel of senior EEGers
Electromagnetic data collected using electroencephalography (EEG) and magnetoencephalography (MEG) are of central importance for psychophysiological research. The scope of concepts, methods, and instruments used by EEG/MEG researchers has dramatically increased and is expected to further increase in the future. Building on existing guideline publications, the goal of the present paper is to contribute to the effective documentation and communication of such advances by providing updated guidelines for conducting and reporting EEG/MEG studies. The guidelines also include a checklist of key information recommended for inclusion in research reports on EEG/MEG measures.
暂无摘要（点击查看原文获取完整内容）
Trial-by-trial variability in perceptual performance on identical stimuli has been related to spontaneous fluctuations in ongoing activity of intrinsic functional connectivity networks (ICNs). In a paradigm requiring sustained vigilance for instance, we previously observed that higher prestimulus activity in a cingulo-insular-thalamic network facilitated subsequent perception. Here, we test our proposed interpretation that this network underpins maintenance of tonic alertness. We used simultaneous acquisition of functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) in the absence of any paradigm to test an ensuing hypothesis, namely that spontaneous fluctuations in this ICN's activity (as measured by fMRI) should show a positive correlation with the electrical signatures of tonic alertness (as recorded by concurrent EEG). We found in human subjects (19 male, 7 female) that activity in a network comprising dorsal anterior cingulate cortex, anterior insula, anterior prefrontal cortex and thalamus is positively correlated with global field power (GFP) of upper alpha band (10-12 Hz) oscillations, the most consistent electrical index of tonic alertness. Conversely, and in line with earlier findings, alpha band power was negatively correlated with activity in another ICN, the so-called dorsal attention network which is most prominently involved in selective spatial attention. We propose that the cingulo-insular-thalamic network serves maintaining tonic alertness through generalized expression of cortical alpha oscillations. Attention is mediated by activity in other systems, e.g., the dorsal attention network for space, selectively disrupts alertness-related suppression and hence manifests as local attenuation of alpha activity.
暂无摘要（点击查看原文获取完整内容）
Quantified electroencephalography (qEEG) was used to compare cerebral electrical variations while human subjects (10 males and 10 females) were observing and executing finger movements and while they were resting. Video recording enabled elimination of subjects performing involuntary movements. EEGs were recorded from 14 sites in seven frequency bands: theta 1, theta 2, alpha 1, alpha, beta 1, beta 2 and beta 3. Analyses were performed on logarithmically transformed absolute spectral power values. Both observation and execution of finger movements involved a decrease in spectral power compared with resting. This decrease was significant only for the alpha 1 frequency band (7.5-10.5 Hz) and it involved nine of the 14 electrode locations (F7, F8, F4, T6, T5, C3, C4, P3 and P4). This indicates that the motor cortex and the frontal cortex are specifically activated by both observation and execution of finger movements. These results provide evidence that observation and execution of movement share the same cortical network.
The Brain Imaging Data Structure (BIDS) project is a rapidly evolving effort in the human brain imaging research community to create standards allowing researchers to readily organize and share study data within and between laboratories. Here we present an extension to BIDS for electroencephalography (EEG) data, EEG-BIDS, along with tools and references to a series of public EEG datasets organized using this new standard.
OBJECTIVE: To define normal and abnormal patterns, test interobserver variability, and the prognostic accuracy of amplitude-integrated electroencephalography (aEEG) soon after the onset of neonatal encephalopathy. METHODS: Consecutive cases of neonatal encephalopathy (n = 56; gestation median, 40; range, 35-42 weeks) and healthy infants (n = 14; gestation median, 40; range, 39-40 weeks) were studied. aEEG was recorded using a cerebral function monitor, at median, 0, range, 0-21 days of age. Of the infants, 24 of the 56 with encephalopathy and all of the normal infants were studied within 12 hours of birth (median, 5; range, 3-12 hours). Forty infants were suspected of having suffered birth asphyxia. Criteria for normal and abnormal patterns were defined and the interobserver variability of these classifications determined. Results were compared with neurodevelopmental outcome assessed at 18 to 24 months of age. aEEG also was compared with a standard EEG and with magnetic resonance imaging. RESULTS: The median upper margin of the widest band of aEEG activity in the control infants was 37.5 microV (range, 30-48 microV), and median lower margin was 8 microV (range, 6.5-11 microV). We classified the aEEG background activity as normal amplitude, the upper margin of band of aEEG activity >10 microV and the lower margin >5 microV; moderately abnormal amplitude, the upper margin of band of aEEG activity >10 microV and the lower margin </=5 microV; and suppressed amplitude, the upper margin of the band of aEEG activity <10 microV and lower margin <5 microV. Recordings were analyzed further for the presence of seizures, defined as periods of sudden increase in voltage accompanied by a narrowing of the band of aEEG activity. Tests of interobserver variability showed excellent agreement both for assessment of amplitude (kappa statistic = 0.85) and for identification of seizures (kappa statistic = 0.76) There was a close relationship between the aEEG and subsequent outcome: 19 of 21 infants with a normal aEEG finding were normal on follow-up at 18 to 24 months of age, whereas 27 of 35 infants with a moderately abnormal or suppressed aEEG and/or seizures died or developed neurologic abnormalities. Thus, aEEG predicted outcome with a sensitivity of 0. 93, a specificity of 0.70, positive predictive value of 0.77, negative predictive value of 0.90, and the likelihood ratio of a positive result of 3.1 and a negative result of 0.06. For the 24 infants studied within 12 hours of birth, the corresponding results were sensitivity, 1.0; specificity, 0.82; positive predictive value, 0.85; negative predictive value, 1; likelihood ratio of a positive result, 5.5; and likelihood ratio of a negative result, 0.18. CONCLUSION: The aEEG is a simple but accurate and reproducible clinical tool that could be useful in the assessment of infants with encephalopathy.
OBJECTIVE: Stereo-electroencephalography (SEEG) is a procedure performed for patients with intractable epilepsy in order to anatomically define the epileptogenic zone (EZ) and the possible related functional cortical areas. By avoiding the need for large craniotomies and due to its intrinsic precision placement features, SEEG may be associated with fewer complications. Nevertheless, intracerebral electrodes have gained a reputation of excessive invasiveness, with a "relatively high morbidity" associated with their placement. A systematic literature review and meta-analysis of SEEG complications has not been previously performed. The goal of this study is to quantitatively review the incidence of various surgical complications associated with SEEG electrode implantation in the literature and to provide a summary estimate. This will allow physicians to accurately counsel their patients about the potential complications related to this method of extraoperative invasive monitoring. METHODS: The systematic review was conducted following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). We conducted MEDLINE, Scopus, and Web of Science database searches with the search algorithm. We analyzed complication rates using a fixed-effects model with inverse variance weighting. Calculations for the meta-analysis and construction of forest plots were completed using an established spreadsheet. The principal summary measures were the effect summary value and 95% confidence intervals (CIs). RESULTS: The initial 1,901 retrieved citations were reviewed. After removing 787 duplicates, the titles and abstracts of 1,114 publications were screened. At this stage, studies that did not mention the absence or presence of complications following SEEG or that did not fulfill the inclusion criteria in any manner were excluded. After excluding 1,057 citations, the full text was assessed in the resulting 57 articles for eligibility criteria. The most common complications were hemorrhagic (pooled prevalence 1.0%, 95% confidence interval [CI] 0.6-1.4%) or infectious (pooled prevalence 0.8%, 95% CI 0.3-1.2%). Five mortalities were identified (pooled prevalence 0.3%, 95% CI -0.1-0.6%). Overall, our analysis identified 121 surgical complications related to SEEG insertion and monitoring (pooled prevalence 1.3%, 95% CI 0.9-1.7%). SIGNIFICANCE: This review represents a comprehensive estimation of the actual incidence of complications related to SEEG. We report a rate substantially lower than the complication rates reported for other methods of extraoperative invasive monitoring. These data should alleviate the concerns of some regarding the safety of the "stereotactic" method, allowing a better decision process among the different methods of invasive monitoring and ameliorating the fear associated with the placement of depth electrodes.
A realistically shaped three-dimensional single-neuron model was constructed for each of four principal cell types in the neocortex in order to infer their contributions to magnetoencephalography (MEG) and electroencephalography (EEG) signals. For each cell, the soma was stimulated and the resulting intracellular current was used to compute the current dipole Q for the whole cell or separately for the apical and basal dendrites. The magnitude of Q is proportional to the magnetic field and electrical potential far from the neuron. A train of spikes and depolarization shift in an intracellular burst discharge were seen as spikes and an envelope in Q for the layer V and layer II/III pyramidal cells. The stellate cells lacked the envelope. As expected, the pyramidal cells produced a stronger Q than the stellate cells. The spikes produced by the layer V pyramidal cells (n = 4) varied between -0.78 and 2.97 pA m with the majority of the cells showing a current toward the pia (defined as positive). The basal dendrites, however, produced considerable spike currents. The magnitude and direction of dipole moment are in agreement with the distribution of the dendrites. The spikes in Q for the layer V pyramidal cells were produced by the transient sodium conductance and potassium conductance of delayed rectifier type; the conductances distributed along the dendrites were capable of generating spike propagation, which was seen in Q as the tail of a triphasic wave lasting several milliseconds. The envelope was similar in magnitude (-0.41 to -0.90 pA m) across the four layer V pyramidal cells. The spike and envelope for the layer II/III pyramidal cell were 0.47 and -0.29 pA m, respectively; these values agreed well with empirical and theoretical estimates for guinea pig CA3 pyramidal cells. Spikes were stronger for the layer IV spiny stellate (0.27 pA m) than the layer III aspiny stellate cell (0.06 pA m) along their best orientations. The spikes may thus be stronger than has been previously thought. The Q for a population of stellate cells may be weaker than a linear sum of their individual Q values due to their variable dendritic geometry. The burst discharge by pyramidal cells may be detectable with MEG and EEG when 10 000-50 000 cells are synchronously active.
OBJECTIVES: To examine predictors and the prognostic value of electrographic seizures (ESZs) and periodic epileptiform discharges (PEDs) in medical intensive care unit (MICU) patients without a primary acute neurologic condition. DESIGN: Retrospective study. SETTING: MICU in a university hospital. PATIENTS: A total of 201 consecutive patients admitted to the MICU between July 2004 and January 2007 without known acute neurologic injury and who underwent continuous electroencephalography monitoring (cEEG) for investigation of possible seizures or changes in mental status. INTERVENTION: None. MEASUREMENTS AND MAIN RESULTS: Median time from intensive care unit (ICU) admission to cEEG was 1 day (interquartile range 1-4). The majority of patients (60%) had sepsis as the primary admission diagnosis and 48% were comatose at the time of cEEG. Ten percent (n = 21) of patients had ESZs, 17% (n = 34) had PEDs, 5% (n = 10) had both, and 22% (n = 45) had either ESZs or PEDs. Seizures during cEEG were purely electrographic (no detectable clinical correlate) in the majority (67%) of patients. Patients with sepsis had a higher rate of ESZs or PEDs than those without sepsis (32% vs. 9%, p < 0.001). On multivariable analysis, sepsis at ICU admission was the only significant predictor of ESZs or PEDs (odds ratio 4.6, 95% confidence interval 1.9-12.7, p = 0.002). After controlling for age, coma, and organ dysfunction, the presence of ESZs or PEDs was associated with death or severe disability at hospital discharge (89% with ESZs or PEDs, vs. 39% if not; odds ratio 19.1, 95% confidence interval 6.3-74.6, p < 0.001). CONCLUSION: In this retrospective study of MICU patients monitored with cEEG, ESZs and PEDs were frequent, predominantly in patients with sepsis. Seizures were mainly nonconvulsive. Both seizures and periodic discharges were associated with poor outcome. Prospective studies are warranted to determine more precisely the frequency and clinical impact of nonconvulsive seizures and periodic discharges, particularly in septic patients.
Human brain consists of millions of neurons which are playing an important role for controlling behavior of human body with respect to internal/external motor/sensory stimuli. These neurons will act as information carriers between human body and brain. Understanding cognitive behaviour of brain can be done by analyzing either signals or images from the brain. Human behaviour can be visualized in terms of motor and sensory states such as, eye movement, lip movement, remembrance, attention, hand clenching etc. These states are related with specific signal frequency which helps to understand functional behavior of complex brain structure. Electroencephalography (EEG) is an efficient modality which helps to acquire brain signals corresponds to various states from the scalp surface area. These signals are generally categorized as delta, theta, alpha, beta and gamma based on signal frequencies ranges from 0.1 Hz to more than 100 Hz. This paper primarily focuses on EEG signals and its characterization with respect to various states of human body. It also deals with experimental setup used in EEG analysis.
This book has 44 chapters written by 23 authors from Europe and the United States. It represents a commendable attempt at producing a one-volume, comprehensive textbook on EEG. Many chapters are good, but some are especially clear and didactic. "Biophysical Aspects of EEG and MEG Generation" deals with its subject in a clear and well-organized fashion understandable to all readers, leaving the mathematical foundations in an appendix for those whose interests are more basic. The chapter on the EEG laboratory goes into a discussion of laboratory design and organization, including comments on reporting EEGs and handling the reports. This is a useful feature that is not commonly seen in texts on EEG. The contingent negative variation is amply discussed in a review that includes techniques and physiologic and psychological correlates, updating previous reviews by the authors of the chapters. Some of the chapters are outstanding. The one on psychiatric disorders
暂无摘要（点击查看原文获取完整内容）
Although magnetoencephalography (MEG) and electroencephalography (EEG) have been available for decades, their relative merits are still debated. We examined regional differences in signal-to-noise-ratios (SNRs) of cortical sources in MEG and EEG. Data from four subjects were used to simulate focal and extended sources located on the cortical surface reconstructed from high-resolution magnetic resonance images. The SNR maps for MEG and EEG were found to be complementary. The SNR of deep sources was larger in EEG than in MEG, whereas the opposite was typically the case for superficial sources. Overall, the SNR maps were more uniform for EEG than for MEG. When using a noise model based on uniformly distributed random sources on the cortex, the SNR in MEG was found to be underestimated, compared with the maps obtained with noise estimated from actual recorded MEG and EEG data. With extended sources, the total area of cortex in which the SNR was higher in EEG than in MEG was larger than with focal sources. Clinically, SNR maps in a patient explained differential sensitivity of MEG and EEG in detecting epileptic activity. Our results emphasize the benefits of recording MEG and EEG simultaneously.