Independent component analysis: comparison of algorithms for the investigation of surface electrical brain activity

EEG microstate as a biomarker of post-stroke depression with acupuncture treatment

Background

Post-stroke depression (PSD) is a prevalent psychiatric complication among stroke survivors. The PSD researches focus on pathogenesis, new treatment methods and efficacy prediction. This study explored the electroencephalography (EEG) microstates in PSD and assessed their changes after acupuncture treatment, aiming to find the biological characteristics and the predictors of treatment efficacy of PSD.

Methods

A 64-channel resting EEG data was collected from 70 PSD patients (PSD group) and 40 healthy controls (HC group) to explore the neuro-electrophysiological mechanism of PSD. The PSD patients received 6 weeks of acupuncture treatment. EEG data was collected from 60 PSD patients after acupuncture treatment (MA group) to verify whether acupuncture had a modulating effect on abnormal EEG microstates. Finally, the MA group was divided into two groups: the remission prediction group (RP group) and the non-remission prediction group (NRP group) according to the 24-Item Hamilton Depression Scale (HAMD-24) reduction rate. A prediction model for acupuncture treatment was established by baseline EEG microstates.

Results

The duration of microstate D along with the occurrence and contribution of microstate C were reduced in PSD patients. Acupuncture treatment partially normalized abnormal EEG microstates in PSD patients. Baseline EEG microstates predicted the efficacy of acupuncture treatment with an area under the curve (AUC) of 0.964.

Conclusion

This study provides a novel viewpoint on the neurophysiological mechanisms of PSD and emphasizes the potential of EEG microstates as a functional biomarker. Additionally, we anticipated the therapeutic outcomes of acupuncture by analyzing the baseline microstates, which holds significant practical implication for the PSD treatment.

EEG artifact removal using sub-space decomposition, nonlinear dynamics, stationary wavelet transform and machine learning algorithms

Blind source separation (BSS) methods have received a great deal of attention in electroencephalogram (EEG) artifact elimination as they are routine and standard signal processing tools to remove artifacts and reserve desired neural information. On the other hand, a classifier should follow BSS methods to automatically identify artifactual sources and remove them in the following steps. In addition, removing all detected artifactual components leads to loss of information since some desired information related to neural activity leaks to these sources. So, an approach should be employed to detect and suppress the artifacts and reserve neural activity. This study introduces a novel method based on EEG and Poincare planes in the phase space to detect artifactual components estimated by second-order blind identification (SOBI). Artifacts are detected using a mixture of well-known conventional classifiers and were removed employing stationary wavelet transform (SWT) to reserve neural information. The proposed method is a combination of signal processing techniques and machine learning algorithms, including multi-layer perceptron (MLP), K-nearest neighbor (KNN), naïve Bayes, and support vector machine (SVM) which have significant results while applying our proposed method to different scenarios. Simulated, semi-simulated, and real EEG signals are employed to evaluate the proposed method, and several evaluation criteria are calculated. We achieved acceptable results, for example, 98% average accuracy and 97% average sensitivity in artifactual EEG component detection or about 2% as mean square error in EEG reconstruction after artifact removal. Results showed that the proposed method is effective and can be used in future studies as we have considered different real-world scenarios to evaluate it.

Comparison of decomposition algorithms for identification of single motor units in ultrafast ultrasound image sequences of low force voluntary skeletal muscle contractions

OBJECTIVE

In this study, the aim was to compare the performance of four spatiotemporal decomposition algorithms (stICA, stJADE, stSOBI, and sPCA) and parameters for identifying single motor units in human skeletal muscle under voluntary isometric contractions in ultrafast ultrasound image sequences as an extension of a previous study. The performance was quantified using two measures: (1) the similarity of components' temporal characteristics against gold standard needle electromyography recordings and (2) the agreement of detected sets of components between the different algorithms.

RESULTS

We found that out of these four algorithms, no algorithm significantly improved the motor unit identification success compared to stICA using spatial information, which was the best together with stSOBI using either spatial or temporal information. Moreover, there was a strong agreement of detected sets of components between the different algorithms. However, stJADE (using temporal information) provided with complementary successful detections. These results suggest that the choice of decomposition algorithm is not critical, but there may be a methodological improvement potential to detect more motor units.

A New Approach for Motor Imagery Classification Based on Sorted Blind Source Separation, Continuous Wavelet Transform, and Convolutional Neural Network

Brain-Computer Interfaces (BCI) are systems that allow the interaction of people and devices on the grounds of brain activity. The noninvasive and most viable way to obtain such information is by using electroencephalography (EEG). However, these signals have a low signal-to-noise ratio, as well as a low spatial resolution. This work proposes a new method built from the combination of a Blind Source Separation (BSS) to obtain estimated independent components, a 2D representation of these component signals using the Continuous Wavelet Transform (CWT), and a classification stage using a Convolutional Neural Network (CNN) approach. A criterion based on the spectral correlation with a Movement Related Independent Component (MRIC) is used to sort the estimated sources by BSS, thus reducing the spatial variance. The experimental results of 94.66% using a k-fold cross validation are competitive with techniques recently reported in the state-of-the-art.

A novel approach to emotion recognition using local subset feature selection and modified Dempster-Shafer theory

Background

Emotion recognition is an increasingly important field of research in brain computer interactions.

Introduction

With the advance of technology, automatic emotion recognition systems no longer seem far-fetched. Be that as it may, detecting neural correlates of emotion has remained a substantial bottleneck. Settling this issue will be a breakthrough of significance in the literature.

Methods

The current study aims to identify the correlations between different emotions and brain regions with the help of suitable electrodes. Initially, independent component analysis algorithm is employed to remove artifacts and extract the independent components. The informative channels are then selected based on the thresholded average activity value for obtained components. Afterwards, effective features are extracted from selected channels common between all emotion classes. Features are reduced using the local subset feature selection method and then fed to a new classification model using modified Dempster-Shafer theory of evidence.

Results

The presented method is employed to DEAP dataset and the results are compared to those of previous studies, which highlights the significant ability of this method to recognize emotions through electroencephalography, by the accuracy of about 91%. Finally, the obtained results are discussed and new aspects are introduced.

Conclusions

The present study addresses the long-standing challenge of finding neural correlates between human emotions and the activated brain regions. Also, we managed to solve uncertainty problem in emotion classification which is one of the most challenging issues in this field. The proposed method could be employed in other practical applications in future.

Removal of muscular artifacts in EEG signals: a comparison of linear decomposition methods

The most common approach to reduce muscle artifacts in electroencephalographic signals is to linearly decompose the signals in order to separate artifactual from neural sources, using one of several variants of independent component analysis (ICA). Here we compare three of the most commonly used ICA methods (extended Infomax, FastICA and TDSEP) with two other linear decomposition methods (Fourier-ICA and spatio-spectral decomposition) suitable for the extraction of oscillatory activity. We evaluate the methods’ ability to remove event-locked muscle artifacts while maintaining event-related desynchronization in data from 18 subjects who performed self-paced foot movements. We find that all five analyzed methods drastically reduce the muscle artifacts. For the three ICA methods, adequately high-pass filtering is very important. Compared to the effect of high-pass filtering, differences between the five analyzed methods were small, with extended Infomax performing best.

Second order blind identification on the cerebral cortex

Blind source separation (BSS) methods have become standard brain imaging tools and are routinely used for noise and artifact removal, as well as for extracting related spatial and temporal components from brain signals. Despite their popularity, BSS methods have rarely been used to explore maps of cortical thickness and sulcal folding patterns. Our limited knowledge of the relationship between cortical morphometry, brain development and pathologies of the central nervous system makes BSS methods ideal investigative tools. We propose a novel spatial BSS method tailored for application to the cerebral cortex based on the second order blind identification (SOBI) method. Our method outperforms the regular SOBI and popular FastICA BSS methods on simulations. Application to maps of cortical thickness and curvature from normal controls reveals original structural networks.

Time-variant partial directed coherence for analysing connectivity: a methodological study

For the past decade, the detection and quantification of interactions within and between physiological networks has become a priority-in-common between the fields of biomedicine and computer science. Prominent examples are the interaction analysis of brain networks and of the cardiovascular-respiratory system. The aim of the study is to show how and to what extent results from time-variant partial directed coherence analysis are influenced by some basic estimator and data parameters. The impacts of the Kalman filter settings, the order of the autoregressive (AR) model, signal-to-noise ratios, filter procedures and volume conduction were investigated. These systematic investigations are based on data derived from simulated connectivity networks and were performed using a Kalman filter approach for the estimation of the time-variant multivariate AR model. Additionally, the influence of electrooculogram artefact rejection on the significance and dynamics of interactions in 29 channel electroencephalography recordings, derived from a photic driving experiment, is demonstrated. For artefact rejection, independent component analysis was used. The study provides rules to correctly apply particular methods that will aid users to achieve more reliable interpretations of the results.

Neural substrate of the late positive potential in emotional processing

The late positive potential (LPP) is a reliable electrophysiological index of emotional perception in humans. Despite years of research, the brain structures that contribute to the generation and modulation of LPP are not well understood. Recording EEG and fMRI simultaneously, and applying a recently proposed single-trial ERP analysis method, we addressed the problem by correlating the single-trial LPP amplitude evoked by affective pictures with the blood oxygen level-dependent (BOLD) activity. Three results were found. First, relative to neutral pictures, pleasant and unpleasant pictures elicited enhanced LPP, as well as heightened BOLD activity in both visual cortices and emotion-processing structures such as amygdala and prefrontal cortex, consistent with previous findings. Second, the LPP amplitude across three picture categories was significantly correlated with BOLD activity in visual cortices, temporal cortices, amygdala, orbitofrontal cortex, and insula. Third, within each picture category, LPP-BOLD coupling revealed category-specific differences. For pleasant pictures, the LPP amplitude was coupled with BOLD in occipitotemporal junction, medial prefrontal cortex, amygdala, and precuneus, whereas for unpleasant pictures significant LPP-BOLD correlation was observed in ventrolateral prefrontal cortex, insula, and posterior cingulate cortex. These results suggest that LPP is generated and modulated by an extensive brain network composed of both cortical and subcortical structures associated with visual and emotional processing and the degree of contribution by each of these structures to the LPP modulation is valence specific.

Coupling between visual alpha oscillations and default mode activity

Although, on average, the magnitude of alpha oscillations (8 to 12 Hz) is decreased in task-relevant cortices during externally oriented attention, its fluctuations have significant consequences, with increased level of alpha associated with decreased level of visual processing and poorer behavioral performance. Functional MRI signals exhibit similar fluctuations. The default mode network (DMN) is on average deactivated in cognitive tasks requiring externally oriented attention. Momentarily insufficient deactivation of DMN, however, is often accompanied by decreased efficiency in stimulus processing, leading to attentional lapses. These observations appear to suggest that visual alpha power and DMN activity may be positively correlated. To what extent such correlation is preserved in the resting state is unclear. We addressed this problem by recording simultaneous EEG-fMRI from healthy human participants under two resting-state conditions: eyes-closed and eyes-open. Short-time visual alpha power was extracted as time series, which was then convolved with a canonical hemodynamic response function (HRF), and correlated with blood-oxygen-level-dependent (BOLD) signals. It was found that visual alpha power was positively correlated with DMN BOLD activity only when the eyes were open; no such correlation existed when the eyes were closed. Functionally, this could be interpreted as indicating that (1) under the eyes-open condition, strong DMN activity is associated with reduced visual cortical excitability, which serves to block external visual input from interfering with introspective mental processing mediated by DMN, while weak DMN activity is associated with increased visual cortical excitability, which helps to facilitate stimulus processing, and (2) under the eyes-closed condition, the lack of external visual input renders such a gating mechanism unnecessary.

Neural substrate of the late positive potential in emotional processing

The late positive potential (LPP) is a reliable electrophysiological index of emotional perception in humans. Despite years of research, the brain structures that contribute to the generation and modulation of LPP are not well understood. Recording EEG and fMRI simultaneously, and applying a recently proposed single-trial ERP analysis method, we addressed the problem by correlating the single-trial LPP amplitude evoked by affective pictures with the blood oxygen level-dependent (BOLD) activity. Three results were found. First, relative to neutral pictures, pleasant and unpleasant pictures elicited enhanced LPP, as well as heightened BOLD activity in both visual cortices and emotion-processing structures such as amygdala and prefrontal cortex, consistent with previous findings. Second, the LPP amplitude across three picture categories was significantly correlated with BOLD activity in visual cortices, temporal cortices, amygdala, orbitofrontal cortex, and insula. Third, within each picture category, LPP-BOLD coupling revealed category-specific differences. For pleasant pictures, the LPP amplitude was coupled with BOLD in occipitotemporal junction, medial prefrontal cortex, amygdala, and precuneus, whereas for unpleasant pictures significant LPP-BOLD correlation was observed in ventrolateral prefrontal cortex, insula, and posterior cingulate cortex. These results suggest that LPP is generated and modulated by an extensive brain network composed of both cortical and subcortical structures associated with visual and emotional processing and the degree of contribution by each of these structures to the LPP modulation is valence specific.

Context modulates the ERP signature of contour integration

We investigated how the electrophysiological signature of contour integration is changed by the context in which a contour is embedded. Specifically, we manipulated the orientations of Gabor elements surrounding an embedded shape outline. The amplitudes of early visual components over posterior scalp regions were changed by the presence of a contour, and by the orientation of elements surrounding the contour. Differences in context type had an effect on the early P1 and N1 components, but not on the later P2 component. The presence of a contour had an effect on the N1 and P2 components, but not on the earlier P1 component. A modulatory effect of context on contour integration was observed on the N1 component. These results highlight the importance of the context in which contour integration takes place.

Electromyogenic Artifacts and Electroencephalographic Inferences Revisited

Recent years have witnessed a renewed interest in using oscillatory brain electrical activity to understand the neural bases of cognition and emotion. Electrical signals originating from pericranial muscles represent a profound threat to the validity of such research. Recently, McMenamin et al (2010) examined whether independent component analysis (ICA) provides a sensitive and specific means of correcting electromyogenic (EMG) artifacts. This report sparked the accompanying commentary (Olbrich, Jödicke, Sander, Himmerich & Hegerl, in press), and here we revisit the question of how EMG can alter inferences drawn from the EEG and what can be done to minimize its pernicious effects. Accordingly, we briefly summarize salient features of the EMG problem and review recent research investigating the utility of ICA for correcting EMG and other artifacts. We then directly address the key concerns articulated by Olbrich and provide a critique of their efforts at validating ICA. We conclude by identifying key areas for future methodological work and offer some practical recommendations for intelligently addressing EMG artifact.

Assessment of artefact suppression by ICA and spatial filtering on reduced sets of EEG signals

In recorded EEG signals, the signal components under interest are typically embedded in noise and artefacts. Independent Component Analysis has been demonstrated to be very successful at signal-to-noise ratio enhancement and artefact suppression, but mainly on a large set of EEG channels (20 or more) and typically on signals from healthy young subjects. In this paper, we assess the artefact suppression performance of five different ICA methods (AMUSE, FASTICA, RUNICA, SOBI and THINICA) combined with four different spatial filters on reduced sets of EEG channels from elderly tremor patients. Results demonstrate that a suitable combination of ICA and spatial filtering can effectively suppress artefacts in clinical EEG signals, even on very small sets with only three EEG channels.

Human ECGs corrupted with real CPR artefacts in an animal model: generating a database to evaluate and refine algorithms for eliminating CPR artefacts

AIM

For the analysis of ECG rhythms during ongoing CPR, single- or two-channel methods have been proposed to eliminate artefacts from the CPR-corrupted ECG. To refine, test and evaluate these algorithms with a realistic data set, we introduce an animal model with which we created an extended database of human ECGs with real CPR artefacts.

MATERIAL AND METHODS

In a pig model real CPR-related artefacts were added to annotated human emergency ECGs. Via a special catheter placed in the oesophagus, ECG sequences (duration>10s) were fed in close to the dead pig's heart. The resulting surface potential was recorded on the thorax without and during ongoing chest compressions, which were monitored using a miniature force sensor.

RESULTS

The animals served as a vehicle for human ECGs, making it possible to create a database in which 918 real human ECG sequences (437 shockable and 481 non-shockable) were corrupted with CPR-induced artefacts. The achieved signal-to-noise ratios (SNR) ranged from -17 to +15 dB, sensitivity was 93.5% and specificity was 50.51%. The fed-in ECG and the uncorrupted surface ECG correlated almost perfectly (r=0.926+/-0.081; n=918), indicating negligible signal distortion due to the dead pig itself.

CONCLUSION

As the generated database includes both the original and the corrupted ECG covering a wide range of SNRs as well as the compression force signal, it provides an extended data set to evaluate the reconstruction performance of CPR artefact-removal algorithms.

Feature selection on movement imagery discrimination and attention detection

Noninvasive brain-computer interfaces (BCI) translate subject's electroencephalogram (EEG) features into device commands. Large feature sets should be down-selected for efficient feature translation. This work proposes two different feature down-selection algorithms for BCI: (a) a sequential forward selection; and (b) an across-group variance. Power rar ratios (PRs) were extracted from the EEG data for movement imagery discrimination. Event-related potentials (ERPs) were employed in the discrimination of cue-evoked responses. While center-out arrows, commonly used in calibration sessions, cued the subjects in the first experiment (for both PR and ERP analyses), less stimulating arrows that were centered in the visual field were employed in the second experiment (for ERP analysis). The proposed algorithms outperformed other three popular feature selection algorithms in movement imagery discrimination. In the first experiment, both algorithms achieved classification errors as low as 12.5% reducing the feature set dimensionality by more than 90%. The classification accuracy of ERPs dropped in the second experiment since centered cues reduced the amplitude of cue-evoked ERPs. The two proposed algorithms effectively reduced feature dimensionality while increasing movement imagery discrimination and detected cue-evoked ERPs that reflect subject attention.

Removal of BCG artifacts from EEG recordings inside the MR scanner: a comparison of methodological and validation-related aspects

Multimodal approaches are of growing interest in the study of neural processes. To this end much attention has been paid to the integration of electroencephalographic (EEG) and functional magnetic resonance imaging (fMRI) data because of their complementary properties. However, the simultaneous acquisition of both types of data causes serious artifacts in the EEG, with amplitudes that may be much larger than those of EEG signals themselves. The most challenging of these artifacts is the ballistocardiogram (BCG) artifact, caused by pulse-related electrode movements inside the magnetic field. Despite numerous efforts to find a suitable approach to remove this artifact, still a considerable discrepancy exists between current EEG-fMRI studies. This paper attempts to clarify several methodological issues regarding the different approaches with an extensive validation based on event-related potentials (ERPs). More specifically, Optimal Basis Set (OBS) and Independent Component Analysis (ICA) based methods were investigated. Their validation was not only performed with measures known from previous studies on the average ERPs, but most attention was focused on task-related measures, including their use on trial-to-trial information. These more detailed validation criteria enabled us to find a clearer distinction between the most widely used cleaning methods. Both OBS and ICA proved to be able to yield equally good results. However, ICA methods needed more parameter tuning, thereby making OBS more robust and easy to use. Moreover, applying OBS prior to ICA can optimize the data quality even more, but caution is recommended since the effect of the additional ICA step may be strongly subject-dependent.