Time-frequency Analysis Of EEG Research Articles

Cycle-frequency content EEG analysis improves the assessment of respiratory-related cortical activity.

Time-Frequency (T-F) analysis of EEG is a common technique to characterise spectral changes in neural activity. This study explores the limitations of utilizing conventional spectral techniques in examining cyclic event-related cortical activities due to challenges, including high inter-trial variability. Introducing the Cycle-Frequency (C-F) analysis, we aim to enhance the evaluation of cycle-locked respiratory events. For synthetic EEG that mimicked cycle-locked pre-motor activity, C-F had more accurate frequency and time localization compared to conventional T-F analysis, even for a significantly reduced number of trials and a variability of breathing rhythm. Preliminary validations using real EEG data during both unloaded breathing and loaded breathing (that evokes pre-motor activity) suggest potential benefits of using the C-F method, particularly in normalizing time units to cyclic activity phases and refining baseline placement and duration. The proposed approach could provide new insights for the study of rhythmic neural activities, complementing T-F analysis.&#xD.

EEG rhythm separation and time-frequency analysis of fast multivariate empirical mode decomposition for motor imagery BCI.

Motor imagery electroencephalogram (EEG) is widely employed in brain-computer interface (BCI) systems. As a time-frequency analysis method for nonlinear and non-stationary signals, multivariate empirical mode decomposition (MEMD) and its noise-assisted version (NA-MEMD) has been widely used in the preprocessing step of BCI systems for separating EEG rhythms corresponding to specific brain activities. However, when applied to multichannel EEG signals, MEMD or NA-MEMD often demonstrate low robustness to noise and high computational complexity. To address these issues, we have explored the advantages of our recently proposed fast multivariate empirical mode decomposition (FMEMD) and its noise-assisted version (NA-FMEMD) for analyzing motor imagery data. We emphasize that FMEMD enables a more accurate estimation of EEG frequency information and exhibits a more noise-robust decomposition performance with improved computational efficiency. Comparative analysis with MEMD on simulation data and real-world EEG validates the above assertions. The joint average frequency measure is employed to automatically select intrinsic mode functions that correspond to specific frequency bands. Thus, FMEMD-based classification architecture is proposed. Using FMEMD as a preprocessing algorithm instead of MEMD can improve the classification accuracy by 2.3% on the BCI Competition IV dataset. On the Physiobank Motor/Mental Imagery dataset and BCI Competition IV Dataset 2a, FMEMD-based architecture also attained a comparable performance to complex algorithms. The results indicate that FMEMD proficiently extracts feature information from small benchmark datasets while mitigating dimensionality constraints resulting from computational complexity. Hence, FMEMD or NA-FMEMD can be a powerful time-frequency preprocessing method for BCI.

Efficient GPU implementation of the multivariate empirical mode decomposition algorithm

An efficient GPU implementation of the Multivariate Empirical Mode Decomposition (MEMD) method is presented for speeding up the process of decomposing non-stationary multi-channel bioelectric signals into different oscillation modes. Each step of the MEMD algorithm is designed with performance in mind and implemented to remove all unnecessary overheads caused by CPU-GPU communication, data transfer operations and synchronisation. The implementation is validated with synthetic and real EEG signals of different lengths and channels (up to 128 channels) on different GPU cards, and compared to existing serial MEMD implementations. The final implementation achieved between 180x-430x speedup compared to MATLAB and a 10x improvement over the only known existing GPU implementation. The average decomposition error of our implementation is below 1.2 %. Our GPU program is the fastest known GPU implementation of the MEMD algorithm that reduces execution time from hours to seconds and as such makes it possible to perform MEMD time-frequency analysis of high-density EEG (MEG) or similar multi-channel signals in a fraction of time and opens the road towards its practical applicability.

GPU Implementation of the Improved CEEMDAN Algorithm for Fast and Efficient EEG Time-Frequency Analysis.

Time-frequency analysis of EEG data is a key step in exploring the internal activities of the human brain. Studying oscillations is an important part of the analysis, as they are thought to provide the underlying mechanism for communication between neural assemblies. Traditional methods of analysis, such as Short-Time FFT and Wavelet Transforms, are not ideal for this task due to the time-frequency uncertainty principle and their reliance on predefined basis functions. Empirical Mode Decomposition and its variants are more suited to this task as they are able to extract the instantaneous frequency and phase information but are too time consuming for practical use. Our aim was to design and develop a massively parallel and performance-optimized GPU implementation of the Improved Complete Ensemble EMD with the Adaptive Noise (CEEMDAN) algorithm that significantly reduces the computational time (from hours to seconds) of such analysis. The resulting GPU program, which is publicly available, was validated against a MATLAB reference implementation and reached over a 260× speedup for actual EEG measurement data, and provided predicted speedups in the range of 3000-8300× for longer measurements when sufficient memory was available. The significance of our research is that this implementation can enable researchers to perform EMD-based EEG analysis routinely, even for high-density EEG measurements. The program is suitable for execution on desktop, cloud, and supercomputer systems and can be the starting point for future large-scale multi-GPU implementations.

Mind-wandering does not always rhyme with proactive functioning! Changes in the temporal dynamics of the mPFC-mediated theta oscillations during moments of mind-wandering

The reduced engagement of the cognitive control network has been documented widely during mind-wandering (MW). However, it remains unknown how MW affects the neural dynamics of cognitive control processes. From this perspective, we explored neural dynamics mediated by the medial prefrontal cortex (mPFC). Their engagement can be both transient (or reactive) and anticipated (or proactive). A total of fortyseven healthy subjects (37 females) were engaged in a long-lasting sustained-attention Go/NoGo task. Subjective probes were used to detect MW episodes. Channel-based EEG time-frequency analysis was performed to measure the theta oscillations, an index of the mPFC activity. The theta oscillations were computed immediately after conflictual NoGo trials to explore the reactive engagement of the mPFC. Proactive control was measured on the Go trials preceded the NoGo. Behaviorally, periods of MW were associated with an increase in errors and in RT variability in comparison to on-task periods. The analysis of the frontal midline theta power (MFθ) revealed that MW periods were associated with lower anticipated/proactive engagement and similar transient/reactive engagement of mPFC-mediated processes. Moreover, the communication between the mPFC and the DLPFC, as revealed by the poorer theta synchronization between these two regions, was also impaired during MW periods. Our results provide new insights about performance impairment during MW. They could be an important step in improving the existing understanding of the altered performances that are reported for some disorders that are known to be associated with excessive MW.

Ultra-high-resolution time-frequency analysis of EEG to characterise brain functional connectivity with the application in Alzheimer’s disease

Objective. This study aims to explore the potential of high-resolution brain functional connectivity based on electroencephalogram, a non-invasive low-cost technique, to be translated into a long-overdue biomarker and a diagnostic method for Alzheimer’s disease (AD). Approach. The paper proposes a novel ultra-high-resolution time-frequency nonlinear cross-spectrum method to construct a promising biomarker of AD pathophysiology. Specifically, using the peak frequency estimated from a revised Hilbert–Huang transformation (RHHT) cross-spectrum as a biomarker, the support vector machine classifier is used to distinguish AD from healthy controls (HCs). Main results. With the combinations of the proposed biomarker and machine learning, we achieved a promising accuracy of 89%. The proposed method performs better than the wavelet cross-spectrum and other functional connectivity measures in the temporal or frequency domain, particularly in the Full, Delta and Alpha bands. Besides, a novel visualisation approach developed from topography is introduced to represent the brain functional connectivity, with which the difference between AD and HCs can be clearly displayed. The interconnections between posterior and other brain regions are obviously affected in AD. Significance. Those findings imply that the proposed RHHT approach could better track dynamic and nonlinear functional connectivity information, paving the way for the development of a novel diagnostic approach.

Time-Frequency Analysis of Scalp EEG With Hilbert-Huang Transform and Deep Learning.

Electroencephalography (EEG) is a brain imaging approach that has been widely used in neuroscience and clinical settings. The conventional EEG analyses usually require pre-defined frequency bands when characterizing neural oscillations and extracting features for classifying EEG signals. However, neural responses are naturally heterogeneous by showing variations in frequency bands of brainwaves and peak frequencies of oscillatory modes across individuals. Fail to account for such variations might result in information loss and classifiers with low accuracy but high variation across individuals. To address these issues, we present a systematic time-frequency analysis approach for analyzing scalp EEG signals. In particular, we propose a data-driven method to compute the subject-specific frequency bands for brain oscillations via Hilbert-Huang Transform, lifting the restriction of using fixed frequency bands for all subjects. Then, we propose two novel metrics to quantify the power and frequency aspects of brainwaves represented by sub-signals decomposed from the EEG signals. The effectiveness of the proposed metrics are tested on two scalp EEG datasets and compared with four commonly used features sets extracted from wavelet and Hilbert-Huang Transform. The validation results show that the proposed metrics are more discriminatory than other features leading to accuracies in the range of 94.93% to 99.84%. Besides classification, the proposed metrics show great potential in quantification of neural oscillations and serving as biomarkers in the neuroscience research.

On the Neuronal Dynamics of Aesthetic Experience: Evidence from Electroencephalographic Oscillatory Dynamics.

Aesthetic experiences have an influence on many aspects of life. Interest in the neural basis of aesthetic experiences has grown rapidly in the past decade, and fMRI studies have identified several brain systems supporting aesthetic experiences. Work on the rapid neuronal dynamics of aesthetic experience, however, is relatively scarce. This study adds to this field by investigating the experience of being aesthetically moved by means of ERP and time-frequency analysis. Participants' EEG was recorded while they viewed a diverse set of artworks and evaluated the extent to which these artworks moved them. Results show that being aesthetically moved is associated with a sustained increase in gamma activity over centroparietal regions. In addition, alpha power over right frontocentral regions was reduced in high- and low-moving images, compared to artworks given intermediate ratings. We interpret the gamma effect as an indication for sustained savoring processes for aesthetically moving artworks compared to aesthetically less-moving artworks. The alpha effect is interpreted as an indication of increased attention for aesthetically salient images. In contrast to previous works, we observed no significant effects in any of the established ERP components, but we did observe effects at latencies longer than 1 sec. We conclude that EEG time-frequency analysis provides useful information on the neuronal dynamics of aesthetic experience.

Working memory impairment in children orphaned by parental HIV/AIDS: An event-related potentials study

ABSTRACT A large body of literature has established that children orphaned by HIV/AIDS (‘AIDS orphans’) face numerous challenges, such as parental death, poverty, disrupted school attendance and stigma. All of these early life adversities can have long-lasting effects on brain function, especially the executive functions. Working memory, as one of the most studied aspects of executive functions, is also reported to be impaired in children with early adversity. However, limited data are available regarding how early life adverse events affect the neural dynamic associated with working memory processing in AIDS orphans. This study applied the electroencephalogram (EEG) technique to investigate the working memory process in 81 AIDS orphans and 62 non-orphan controls with n-back tasks. Results from EEG analysis and time-frequency analysis showed that AIDS orphans displayed smaller N2 and larger P2, P3 activation as well as enhanced theta and attenuated alpha band oscillations compared to the controls. The present findings indicated a deficit in working memory process in AIDS orphans and suggested that this deficit might be due to the impairments in attention allocation, detection and classification of stimuli and updating process in working memory.

Insomnia disorder and hyperarousal: evidence from resting-state and sleeping EEG

Insomnia has become the second highly prevalent mental illness, secondary only to depression. The hyperarousal model of insomnia is one of the theories dedicated to clarifying the pathophysiological mechanism of insomnia disorder. The resting-state and sleeping EEG provide substantial evidence for this model. The current review firstly summarizes the analytic methods of the resting-state and sleeping EEG, and these methods can be mainly used to analyze the EEG data during wakefulness, sleep onset period and all sleep stages. Through literature review, we found that the EEG evidence of hyperarousal in insomniacs included impaired sleep continuity and architecture, enhanced frequency of arousal, delayed daytime sleep latency, and increased β activity during wakefulness and non-rapid eye movement sleep. Some interventions, such as cognitive behavioral therapy for insomnia, non-benzodiazepine hypnotics and slow oscillating transcranial direct current stimulation during sleep, turned out to be the effective therapies that could reduce the cortical hyperarousal in patients with insomnia. Power spectral analysis might aid in differentiating some insomnia subtypes, and could be used in the diagnosis and evaluation of the treatment response in patients with insomnia, but the confounding factors such as age, gender, and EEG band subdivision, should be considered. Future studies should focus on different insomnia subtypes, adopt a unified EEG band subdivision, and consider the possible interferences of patients’ age and gender. Based on the polysomnography derived sleep scoring, some meso/micro-analyses such as EEG source imaging and EEG time-frequency analysis should also be our primary concerns.

Functional Connectivity Between Motor and Mid-Frontal Areas During Vicarious Reward Revealed via EEG Time-Frequency Analysis.

Vicarious reward is a phenomenon in which an individual feels as if he/she has received a reward as the result of watching someone else receive a reward. In this study, we used electroencephalography to investigate brain activity while participants watched a preferred player win a competitive game (Rock-Paper-Scissors game). In the experimental task, movie clips showed right hand of the two players and played Rock-Paper-Scissors game. We asked participants to explicitly support or “cheer” for a specific player, and then examined brain activity associated with vicarious reward. For the observed hand movement, previous findings showed that the event-related desynchronization of mu band (8–14 Hz) appeared at the contra-lateral central electrode to the observed hand (If someone sees the right-hand movement, the left central electrode shows the event-related desynchronization of mu-band). During observation of the player, we detected event-related desynchronization of mu band activity in the contra-lateral central electrode as well as mid-frontal beta band (15–22 Hz) activation when the preferred player won. Furthermore, functional connectivity analysis revealed a strong phase synchronization between the contra-lateral central electrode and mid-frontal electrode in the mu band when participants received the vicarious reward. Cross-frequency coupling analysis revealed functional integration between the mu and beta bands at mid-frontal electrode. These results indicate the interaction of mu band observed at contra-lateral electrode and beta band observed at mid-frontal electrode coupling, suggesting a link between the mirror neuron system and the reward system during vicarious reward.

Redundancy Removed Dual-Tree Discrete Wavelet Transform to Construct Compact Representations for Automated Seizure Detection

With the development of pervasive sensing and machine learning technologies, automated epileptic seizure detection based on electroencephalogram (EEG) signals has provided tremendous support for the lives of epileptic patients. Discrete wavelet transform (DWT) is an effective method for time-frequency analysis of EEG and has been used for seizure detection in daily healthcare monitoring systems. However, the shift variance, the lack of directionality and the substantial aliasing, limit the effects of DWT in some applications. Dual-tree discrete wavelet transform (DTDWT) can overcome those drawbacks but may increase information redundancy. For classification tasks with small dataset sizes, such redundancy can greatly reduce learning efficiency and model performance. In this work, we proposed a novel redundancy removed DTDWT (RR-DTDWT) framework for automated seizure detection. Energy and modified multi-scale entropy (MMSE) features in a dual tree wavelet domain were extracted to construct a complete picture of mental states. To the best of our knowledge, this is the first study to employ MMSE as an indicator of epileptic seizures. Moreover, a compact EEG representation can be obtained after removing useless information redundancy (redundancy between wavelet trees, adjacent channels and entropy scales) by a general auto-weighted feature selection framework via global redundancy minimization (AGRM). Through validation on Bonn and CHB-MIT databases, the proposed RR-DTDWT method can achieve better performance than previous studies.

Extinction-resistant attention to long-term conditioned threat is indexed by selective visuocortical alpha suppression in humans

Previous electrophysiological studies in humans have shown rapid modulations of visual attention after conditioned threat vs. safety cues (<500 ms post-stimulus), but it is unknown whether this attentional prioritization is sustained throughout later time windows and whether it is robust to extinction. To investigate sustained visual attention, we assessed visuocortical alpha suppression in response to conditioned and extinguished threat. We reanalysed data from N = 87 male participants that had shown successful long-term threat conditioning and extinction in self reports and physiological measures in a two-day conditioning paradigm. The current EEG time-frequency analyses on recall test data on Day 2 revealed that previously threat-conditioned vs. safety cues evoked stronger occipital alpha power suppression from 600 to 1200 ms. Notably, this suppression was resistant to previous extinction. The present study showed for the first time that threat conditioning enhances sustained modulation of visuocortical attention to threat in the long term. Long-term stability and extinction resistance of alpha suppression suggest a crucial role of visuocortical attention mechanisms in the maintenance of learned fears.

Synchronous brain networks for passive auditory perception in depressive states: A pilot study

Recent studies have revealed a strong relationship between the default mode network (DMN) and major depression disorder (MDD). The DMN consists of several areas in the brain where activity simultaneously increases during the resting state and is suppressed during cognitive tasks (i.e., DMN suppression). Although the DMN has been evaluated in patients with MDD, it has not been studied in people with self-measured depressive symptoms without medication. Although most studies have used high-demand cognitive tasks, the relationships between MDD and passive sensory tasks remain unclear. Here, we recorded electroencephalograph (EEG) data under two sessions: a resting session and an auditory session. Moreover, we assessed depressive states with a Self-Rating Depression Scale (SDS) score. To reveal the DMN suppression mechanism in the depressive states, we used EEG time-frequency analysis. As a result, the alpha-band phase synchronization in the DMN increased during the resting session and decreased during the auditory session. The results suggest that participants in a depressive state have both an abnormal DMN connectivity and a suppressed DMN connectivity via a passive stimulus. Moreover, we were able to estimate the DMN suppression mechanism during the depressive states: (1) the beta-band phase resetting was found in the auditory and parietal areas via the auditory stimulus; (2) the beta-band transfer entropy from the auditory area to the parietal area was high as information flow among these area; and (3) the beta-band systems (information flow) were synchronized with the alpha-band DMN systems. Although the sample size was small, these results suggest that the DMN systems may already be altered during self-measured depressive symptoms like the early stages of the depressive states.

EEG time-frequency analysis provides arguments for arm swing support in human gait control

Background: Human gait benefits from arm swing, which requires four-limb co-ordination. The Supplementary Motor Area (SMA) is involved in multi-limb coordination. With its location anterior to the leg motor cortex and the pattern of its connections, this suggests a distinct role in gait control.Research question: Is the SMA functionally implicated in gait-related arm swing?Methods: Ambulant electroencephalography (EEG) was employed during walking with and without arm swing in twenty healthy subjects (mean age: 64.9yrs, SD 7.2). Power changes across the EEG frequency spectrum were assessed by Event Related Spectral Perturbation (ERSP) analysis over both the putative SMA at electrode position Fz and additional sensorimotor regions.Results: During walking with arm swing, midline electrodes Fz and Cz showed a step-related pattern of Event Related Desynchronization (ERD) followed by Event Related Synchronization (ERS). Walking without arm swing was associated with significant ERD-ERS power reduction in the high-beta/low-gamma band over Fz and a power increase over Cz. Electrodes C3 and C4 revealed a pattern of ERD during contralateral- and ERS during ipsilateral leg swing. This ERD power decreased in gait without arm swing (low-frequency band). The ERSP pattern during walking with arm swing was similar at CP1 and CP2: ERD was seen during double support and the initial swing phase of the right leg, while a strong ERS emerged during the second half of the left leg’s swing. Walking without arm swing showed a significant power reduction of this ERD-ERS pattern over CP2, while over CP1, ERS during left leg’s swing turned into ERD.Conclusion: The relation between arm swing in walking and a step-related ERD-ERS pattern in the high-beta/low-gamma band over the putative SMA, points at an SMA contribution to integrated cyclic anti-phase movements of upper- and lower limbs. This supports a cortical underpinning of arm swing support in gait control.

Pain Expressions and Inhibitory Control in Patients With Fibromyalgia: Behavioral and Neural Correlates.

Fibromyalgia (FM) is a generalized chronic pain condition associated with a variety of symptoms, including altered cognitive and emotional processing. It has been proposed that FM patients show a preferential allocation of attention to information related to the symptoms of the disease, particularly to pain cues. However, the existing literature does not provide conclusive evidence on the presence of this attentional bias, and its effect on cognitive functions such as inhibitory control. To clarify this issue, we recorded the electroencephalographic activity of 31 women diagnosed with FM and 28 healthy women, while performing an emotional Go/NoGo task with micro-videos of pain, happy, and neutral facial expressions. We analyzed behavioral data, performed EEG time-frequency analyses, and obtained the event-related potentials (ERPs) N2 and P3 components in NoGo trials. A series of self-reports was also administered to evaluate catastrophic thinking and the main symptoms of fibromyalgia. Pain expressions were associated with longer reaction times and more errors, as well as with higher theta and delta power, and P3 amplitude to NoGo stimuli. Thus, behavioral and psychophysiological data suggest that increased attention to pain expressions impairs the performance of an inhibitory task, although this effect was similar in FM patients and healthy controls. N2 amplitude was modulated by type of facial expression (larger to pain faces), but only for the control group. This finding suggests that the presentation of pain faces might represent a smaller conflict for the patients, more used to encounter pain stimuli. No main group effects were found significant for N2 or P3 amplitudes, nor for time-frequency data. Using stimuli with greater ecological validity than in previous studies, we could not confirm a greater effect of attentional bias toward negative stimuli over inhibitory performance in patients with FM. Studying these effects allow us to better understand the mechanisms that maintain pain and develop intervention strategies to modify them.

Developmental differences in the neural oscillations underlying auditory sentence processing in children and adults

Although very young children seem to process ongoing language quickly and effortlessly, neuroimaging and behavioral studies reveal that children continue to mature in their language skills through adolescence. During this prolonged development, children likely engage the same basic cognitive processes and neural mechanisms to perform language tasks as adults, but in somewhat different ways. In this study we used time frequency analysis of EEG to identify developmental differences in the engagement of neural oscillations between children (ages 10–12) and adults while listening to naturally-paced sentences. Adults displayed consistent beta changes throughout the sentence compared to children, thought to be related to efficient syntactic integration, and children displayed more broadly distributed theta changes than adults, thought to be related to more effortful semantic integration. Few differences in alpha, related to verbal working memory, existed between groups. These findings shed new light on developmental changes in the neuronal processes underlying language comprehension.

Brain dynamics of decision-making in the generalized trust game: Evidence from ERPs and EEG time-frequency analysis

Brain dynamics of decision-making in the generalized trust game: Evidence from ERPs and EEG time-frequency analysis

Cognitive strategies in the mental rotation task revealed by EEG spectral power

The classic mental rotation task (MRT; Shepard & Metzler, 1971) is commonly thought to measure mental rotation, a cognitive process involving covert simulation of motor rotation. Yet much research suggests that the MRT recruits both motor simulation and other analytic cognitive strategies that depend on visuospatial representation and visual working memory (WM). In the present study, we investigated cognitive strategies in the MRT using time-frequency analysis of EEG and independent component analysis. We scrutinized sensorimotor mu (µ) power reduction, associated with motor simulation, parietal alpha (pα) power reduction, associated with visuospatial representation, and frontal midline theta (fmθ) power enhancement, associated with WM maintenance and manipulation. µ power increased concomitant with increasing task difficulty, suggesting reduced use of motor simulation, while pα decreased and fmθ power increased, suggesting heightened use of visuospatial representation processing and WM, respectively. These findings suggest that MRT performance involves flexibly trading off between cognitive strategies, namely a motor simulation-based mental rotation strategy and WM-intensive analytic strategies based on task difficulty. Flexible cognitive strategy use may be a domain-general cognitive principle that underlies aptitude and spatial intelligence in a variety of cognitive domains. We close with discussion of the present study’s implications as well as future directions.

Theta- and delta-band EEG network dynamics during a novelty oddball task.

While the P3 component during target detection and novelty processing has been widely studied, less is known about its underlying network dynamics. A recent cognitive model suggests that frontal-parietal and frontal-temporal interregional connectivity are related to attention/action selection and target-related memory updating during the P3, respectively, but empirical work testing this model is lacking. Other work suggests the importance of theta- and delta-band connectivity between the medial frontal cortex and distributed cortical regions during attention, stimulus detection, and response selection processes, and similar dynamics may underlie P3-related network connectivity. The present study evaluated the functional connectivity elicited during a visual task, which combined oddball target and novelty stimuli, in a sample of 231 same-sex twins. It was hypothesized that both target and novel conditions would involve theta frontoparietal connectivity and medial frontal theta power, but that target stimuli would elicit the strongest frontotemporal connectivity. EEG time-frequency analysis revealed greater theta-band frontoparietal connectivity and medial frontal power during both target and novel conditions compared to standards, which may index conflict/uncertainty resolution processes. Theta-band frontotemporal connectivity was maximal during the target condition, potentially reflecting context updating or stimulus-response activation. Delta-band frontocentral-parietal connectivity was also strongest following targets, which may be sensitive to response-related demands. These results suggest the existence of functional networks related to P3 that are differentially engaged by target oddballs and novel distractors. Compared to simple P3 amplitude, network measures may provide a more nuanced view of the neural dynamics during target detection/novelty processing in normative and pathological populations.