Electroencephalography (EEG) signals are disrupted by technical and physiological artifacts. One of the most common artifacts is the natural activity that results from the movement of the eyes and the blinking of the subject. Eye blink artifacts (EB) spread across the entire head surface and make EEG signal analysis difficult. Methods for the elimination of electrooculography (EOG) artifacts, such as independent component analysis (ICA) and regression, are known. The aim of this article was to implement the convolutional neural network (CNN) to eliminate eye blink artifacts. To train the CNN, a method for augmenting EEG signals was proposed. The results obtained from the CNN were compared with the results of the ICA and regression methods for the generated and real EEG signals. The results obtained indicate a much better performance of the CNN in the task of removing eye-blink artifacts, in particular for the electrodes located in the central part of the head.
Electroencephalography (EEG) is a widely used cerebral activity measuring device for both clinical and everyday life applications. In addition to denoising and potential classification, a crucial step in EEG processing is to extract relevant features. Topological data analysis (TDA) as an emerging tool enables to analyse and understand data from a different angle than traditionally used methods. As a higher dimensional analogy of graph analysis, TDA can model rich interactions beyond pairwise relations. It also distinguishes different dynamics of EEG time series. TDA remains largely unknown to the EEG processing community while it fits well the heterogeneous nature of EEG signals. This short review aims to give a quick introduction to TDA and how it can be applied to EEG analysis in various applications including brain-computer interfaces (BCIs). After introducing the objective of the article, the main concepts and ideas of TDA are explained. Next, how to implement it for EEG processing is detailed, and lastly the article discusses the benefits and limitations of the method.
Resting-state neural oscillations are used as biomarkers for functional diseases such as dementia, epilepsy, and stroke. However, accurate interpretation of clinical outcomes requires the identification and minimisation of potential confounding factors. While several studies have indicated that the menstrual cycle also alters brain activity, most of these studies were based on visual inspection rather than objective quantitative measures. In the present study, we aimed to clarify the effect of the menstrual cycle on spontaneous neural oscillations based on quantitative magnetoencephalography (MEG) parameters. Resting-state MEG activity was recorded from 25 healthy women with normal menstrual cycles. For each woman, resting-state brain activity was acquired twice using MEG: once during their menstrual period (MP) and once outside of this period (OP). Our results indicated that the median frequency and peak alpha frequency of the power spectrum were low, whereas Shannon spectral entropy was high, during the MP. Theta intensity within the right temporal cortex and right limbic system was significantly lower during the MP than during the OP. High gamma intensity in the left parietal cortex was also significantly lower during the MP than during the OP. Similar differences were also observed in the parietal and occipital regions between the proliferative (the late part of the follicular phase) and secretory phases (luteal phase). Our findings suggest that the menstrual cycle should be considered to ensure accurate interpretation of functional neuroimaging in clinical practice.
Background: While research has consistently identified an association between long-term cannabis use and memory impairments, few studies have examined this relationship in a polydrug context (i.e., when combining cannabis with other substances).
Aims: In this preliminary study, we used event-related potentials to examine the recognition process in a visual episodic memory task in cannabis users (CU) and cannabis polydrug users (PU). We hypothesized that CU and PU will have both–behavioral and psychophysiological–indicators of memory processes affected, compared to matched non-using controls with the PU expressing more severe changes.
Methods: 29 non-using controls (CG), 24 CU and 27 PU were enrolled into the study. All participants completed a visual learning recognition task while brain electrical activity was recorded. Event-related potentials were calculated for familiar (old) and new images from a signal recorded during a subsequent recognition test. We used receiver operating characteristic curves for behavioral data analysis.
Results: The groups did not differ in memory performance based on receiver operating characteristic method in accuracy and discriminability indicators nor mean reaction times for old/new images. The frontal old/new effect expected from prior research was observed for all participants, while a parietal old/new effect was not observed. While, the significant differences in the late parietal component (LPC) amplitude was observed between CG and PU but not between CG and CU nor CU and PU. Linear regression analysis was used to examine the mean amplitude of the LPC component as a predictor of memory performance accuracy indicator. LPC amplitude predicts recognition accuracy only in the CG.
Conclusion: The results showed alterations in recognition memory processing in CU and PU groups compared to CG, which were not manifested on the behavioral level, and were the most prominent in cannabis polydrug users. We interpret it as a manifestation of the cumulative effect of multiple drug usage in the PU group.