Figure 1 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
The selected electrode locations of the International 10–20 system (29 EEG recording electrodes (black circles), one ground and one reference electrode (red circles) used in this paper).
Source publication
Event-related potential (ERP)-based P300 spellers are commonly used in the field of brain-computer interfaces as an alternative channel of communication for people with severe neuro-muscular diseases. This study introduces a novel P300 based brain-computer interface (BCI) stimulus paradigm using a random set presentation pattern and exploiting the...
Similar publications
En este estudio se presenta un navegador web controlado mediante un sistema Brain-Computer Interface (BCI) orientado a personas con grave discapacidad. Se emplean los potenciales evocados P300 del usuario, generados a través de un paradigma odd-ball visual, como señales de control de la aplicación. Hasta donde tenemos conocimiento, todos los navega...
We study the feasibility of a hybrid Brain-Computer Interface (BCI) combining simultaneous visual oddball and Steady-State Visually Evoked Potential (SSVEP) paradigms, where both types of stimuli are superimposed on a computer screen. Potentially, such a combination could result in a system being able to operate faster than a purely P300-based BCI...
The performance of the event-related potential (ERP)-based brain–computer interface (BCI) declines when applying it into the real environment, which limits the generality of the BCI. The sound is a common noise in daily life, and whether it has influence on this decline is unknown. This study designs a visual-auditory BCI task that requires the sub...
The translation of non-invasive Brain-Computer Interfaces (BCI) from the only possible communication pathway for paralyzed patients into more widespread applications is limited by: the lack of effective, user-friendly and robust paradigms, low information transfer rates (ITR), and lack of suitable electroencephalography (EEG) recording platforms. T...
Brain-computer interface (BCI) systems were developed so that people can control computers or machines through their brain activity without moving their limbs. The P300 speller is one of the BCI applications used most commonly, as is very simple and reliable and can achieve satisfactory performance. However, like other BCIs, the P300 speller still...
Citations
... Trial-wise averaging is a primary strategy used to improve the signalto-noise ratio (SNR). Having better stimulus/feedback types [6,7] to modulate a clearer signal and using advanced signal processing methods [8,9] for a more informative feature or better classification are also good approaches to attempt. However, several factors, including physiological and anatomical differences between people, environmental differences, and hardware characteristics, influence EEG [10]. ...
... The paradigm of the Speller-1 and Speller-2 datasets designed the symbols (A to Z, 1-9, and _) with six rows and six columns, and each row and column blinked. However, in the Speller-face dataset's paradigm, rows and columns blinked irrespectively and randomly [6], and the subject was also shown a face image rather than the symbols when the visual stimulus blinked [33]. The EEG data were recorded using BrainAmp (Brain Product, Inc.), which was composed of 62 electrodes at a rate of 1000 Hz, nasion-referenced, and grounded to electrode AFz. ...
... The other interesting point is that the model trained using the Speller-1 or 2 datasets worked well on the Speller-face dataset with the SA method but not in the opposite direction. The speller datasets have the same paradigm, but the Speller-face dataset has certain variations, such as random-set presentation and face blinking [6,33]. Due to the difference in stimulus type, the ERP pattern may differ. ...
Objective. Transfer learning has become an important issue in the brain-computer interface (BCI) field, and studies on subject-to-subject transfer within the same dataset have been performed. However, few studies have been performed on dataset-to-dataset transfer, including paradigm-to-paradigm transfer. In this study, we propose a signal alignment for P300 event-related potential (ERP) signals that is intuitive, simple, computationally less expensive, and can be used for cross-dataset transfer learning.
Approach. We proposed a linear signal alignment that uses the P300’s latency, amplitude scale, and reverse factor to transform signals. For evaluation, four datasets were introduced (two from conventional P300 Speller BCIs, one from a P300 Speller with face stimuli, and the last from a standard auditory oddball paradigm).
Results. Although the standard approach without signal alignment had an average precision score of 25.5%, the approach demonstrated a 35.8% average precision score, and we observed that the number of subjects showing improvement was 36.0% on average. Particularly, we confirmed that the Speller dataset with face stimuli was more comparable with other datasets.
Significance. We proposed a simple and intuitive way to align ERP signals that uses the characteristics of ERP signals. The results demonstrated the feasibility of cross-dataset transfer learning even between datasets with different paradigms.
... In ERP spellers paradigm with grid layout and row/column intensification scheme, most of the detection errors happens in the neighboring icons to the target, a phenomenon called adjacency-distraction errors (Fazel-Rezai 2007). This issue was diminished by a novel stimulus intensification scheme presented in (Townsend et al 2012), however it remains difficult to solve for single-trial setting (Yeom et al 2014). In addition, presenting consecutive visual stimuli may reveal demanding for users when the trial lasts long, which requires fast stimulation. ...
Hybrid Brain-Computer Interface (hBCI) combines multiple neurophysiology
modalities or paradigms to speed up the output of a single command or produce
multiple ones simultaneously. Concurrent hBCIs that employ endogenous and exogenous
paradigms are limited by the reduced set of possible commands. Conversely, the
fusion of different exogenous visual evoked potentials demonstrated impressive performances;
however, they suffer from limited portability. Yet, sequential hBCIs did
not receive much attention mainly due to slower transfer rate and user fatigue high
potential during prolonged BCI use (Lorenz, Pascual, Blankertz & Vidaurre 2014).
Moreover, the crucial factors for optimizing the hybridization remain under-explored.
In this paper, we test the feasibility of sequential Event Related-Potentials (ERP)
and Steady-State Visual Evoked Potentials (SSVEP) hBCI and study the effect of
stimulus order presentation between ERP-SSVEP and SSVEP-ERP for the control of
directions and speed of powered wheelchairs or mobile robots with 15 commands.
Exploiting the fast single trial face stimulus ERP, SSVEP and modern efficient
convolutional neural networks, the configuration with SSVEP presented at first
achieved significantly (p < 0.05) higher average accuracy rate with 76.39% (± 7.30
standard deviation) hybrid command accuracy and an average Information Transfer
Rate (ITR) of 25.05 (± 5.32 standard deviation) bits per minute (bpm).
The results of the study demonstrate the suitability of a sequential SSVEP-ERP
hBCI with challenging dry electroencephalography (EEG) electrodes and low-compute
capacity. Although it presents lower ITR than concurrent hBCIs, our system presents
an alternative in small screen settings when the conditions for concurrent hBCIs are
hardly satisfied.
... One-way ANOVA showed that PITR and TITR were significantly different between the two and three repetitions, however, there is no significant difference in accuracy and capacity C between the two and three repetitions (accuracy: p > 0.05; PITR: p < 0.00001; TITR: p < 0.00001; C: p > 0.05). The performance comparison based on accuracy, PITR, and TITR between the proposed GR-MINMAX-MDRM with the earlier reported techniques [3], [4], [7], [28], [38], [46], [75], [76], [77], [78], [79], [80] is shown in Table 8. ...
A P300-based speller BCI with large target symbols was designed to improve the detection accuracy and information transfer rate (ITR) as well as to overcome adjacency, crowding, and fatigue problems. The speller paradigm consists of a 8×7 speller matrix which uses only 8 flashing characters for selection. Another critical issue in the design of the BCI speller is the feature extraction and classification algorithms. The symmetric positive definite (SPD) matrix of data has emerged as a compact and informative representation of brain signals conveying spatial information. A geometric way to analyze SPD matrices is through the Riemannian geometry (RG) which has been shown to overcome the limitations of Euclidean geometry and provides robustness with respect to outlier and possess good generalization abilities. However, for large data sets, when the number of channels is large, we encounter the curse of dimensionality problem in computing the covariance matrix and constructing the Riemannian graph. To overcome this problem and be able to construct a low-dimensional SPD, we propose a hybrid method based on a convolutional neural network and RG. A mapping that relies on the RG, is proposed to extract the maximum discriminative features. The experimental results from 6 subjects showed that the proposed BCI speller achieved promising accuracy (ITR) of 95±3.6% (26.73±1.8 bits/min) in offline analyses and 94.3±3.8% (54.6 ± 3.8 bits/min) in online experiments. Additionally, the proposed feature extraction method was applied to the BCI competition datasets and compared with most of the state-of-the-art algorithms. The results demonstrated that the accuracy (ITR) of the proposed method was 100% (8.9 bits/min) and 97% (8.5 bits/min) on dataset II from BCI-competition II and dataset II from BCI-competition III, respectively.
... These EEG signals help computer devices to recognize user instructions. Furthermore, varied multiple kind of EEG-based techniques are available with regards to brain-computer communication which are mainly focused on acquisition of brain signals and functioning of brain responses like sensorimotor rhythms [3]- [8], visual evoked potentials [9], motor imagery (MI) [10] and event-related potentials (ERPs) [11] and so on. Here, visual evoked potential (VEPs) are expressed as the brain response modulation, which take place due to visual stimuli in the cortex area. ...
Brain-computer interfaces (BCIs) system is a link to generate a communication between disable people and physical devices. Thus, steady state visually evoked potential (SSVEP) is analysed to improve performance efficiency of BCIs system using multi-class classification process. Thus, an adaptive filtering-based component analysis (AFCA) method is adopted to examine SSVEP from multiple-channel electroencephalography (EEG) signals for BCIs system efficiency enhancement. Further, flickering at varied frequencies is used in a visual stimulation process to examine user intentions and brain responses. A detailed solution for optimization problem and efficient feature extraction is also presented. Here, a large SSVEP dataset is utilized which contains 256 channel EEG data. Experimental results are evaluated in terms of classification accuracy and information transfer rate to measure efficiency of proposed SSVEP extraction method against varied traditional SSVEP-based BCIs. The average information transfer rate (ITR) results are 308.23 bits per minute and classification accuracy is 93.48% using proposed AFCA method. Thus, proposed AFCA method shows decent performance in comparison with state-of-art-SSVEP extraction methods.
... To monitor the activities of the brain, electroencephalography (EEG) is widely used as it provides a huge amount of both physiological and pathological information, thereby proving its validity for an effective diagnosis (Kim et al., 2014;Jukic et al., 2020). The application of EEG plays a vital role in almost all areas of biomedical engineering, ranging from seizure classification (Rajaguru and Prabhakar, 2016), Alzheimer's disease diagnosis (Zhu et al., 2016), subject-dependent classification in Brain-Computer Interface (BCI) , classification of steadystate visual evoked potentials (Won et al., 2015), efficient analysis of Event-Related Potentials (ERP)-based BCI (Yeom et al., 2014), analysis of deep neural network modeling under an ambulatory environment (Kwak et al., 2017), dementia classification (Jeong, 2004), and iris recognition (Adamovic et al., 2020). Visual analysis of EEG recordings, even by expert neurologists, is very difficult and time-consuming. ...
A comprehensive analysis of an automated system for epileptic seizure detection is explained in this work. When a seizure occurs, it is quite difficult to differentiate the non-stationary patterns from the discharges occurring in a rhythmic manner. The proposed approach deals with it efficiently by clustering it initially for the sake of feature extraction by using six different techniques categorized under two different methods, e.g., bio-inspired clustering and learning-based clustering. Learning-based clustering includes K-means clusters and Fuzzy C-means (FCM) clusters, while bio-inspired clusters include Cuckoo search clusters, Dragonfly clusters, Firefly clusters, and Modified Firefly clusters. Clustered values were then classified with 10 suitable classifiers, and after the performance comparison analysis of the EEG time series, the results proved that this methodology flow achieved a good performance index and a high classification accuracy. A comparatively higher classification accuracy of 99.48% was achieved when Cuckoo search clusters were utilized with linear support vector machines (SVM) for epilepsy detection. A high classification accuracy of 98.96% was obtained when K-means clusters were classified with a naive Bayesian classifier (NBC) and Linear SVM, and similar results were obtained when FCM clusters were classified with Decision Trees yielding the same values. The comparatively lowest classification accuracy, at 75.5%, was obtained when Dragonfly clusters were classified with the K-nearest neighbor (KNN) classifier, and the second lowest classification accuracy of 75.75% was obtained when Firefly clusters were classified with NBC.
... With the advancement of machine learning techniques like DL etc, more efficient algorithms have been put forward to classify EEG signals. In [172], authors have used three CNNs to recognize EEG for P300 BCI. But DL algorithms also suffer from certain limitations like the requirement of an excessive quantity of training samples. ...
Small availability of electroencephalograph (EEG) data makes the training of Brain-Computer Interface (BCI) significantly a difficult task. Recently, Deep Learning (DL) has shown tremendous performance in all domains but it requires an abundant amount of data for the training. Another challenging task is the non-stationary nature of EEG data. In this study, we illustrate a roadmap of various adversarial networks, review Transfer Learning (TL) methods, and give insights on the development route to generate and classify motor imagery (MI) EEG data. To make the use of BCI more practical, a Comprehensive review of Generative Adversarial Networks (GANs) and TL was performed to address the following questions: (1) Why GANs, rather than expansion on traditional data augmentation technique? (2) What input formulations have been used for generating brain signal data using GANs? (3) Are there particular GANs suitable for EEG data? (4) How can we solve the problem of data scarcity and the non-stationary nature of EEG? Comprehensive literature on data augmentation using GANs and feature transferring using advanced TL has been performed. The studies were examined based on different GAN architectures, various input formulations of EEG data, and different advanced TL techniques. For MI EEG data augmentation, signal input formulation is classified into five types and advanced TL techniques into seven types. For all these categories, the previous work from a technical point of view is discussed. Towards the end of the paper, a novelistic hybrid approach based on data augmentation and TL has been put forward for classifying MI EEG data with a short-term calibration process. To the best of our knowledge, no review article has completely discussed the various variants of GANs for augmenting brain signals. This review article will help the researchers in the deployment of BCI in future research.
... The 10-20 electrode system of the International Federation[12]. ...
Driver fatigue reduces the safety of traditional driving and limits the widespread adoption of self-driving cars; hence, the monitoring and early detection of drivers’ drowsiness plays a key role in driving automation. When representing the drowsiness indicators as large feature vectors, fitting a machine learning model to the problem becomes challenging, and the problem’s perspicuity decreases, making dimensionality reduction crucial in practice. For this reason, we propose an embedded feature selection algorithm that can be later utilized as a building block in the system development of a neural network-based drowsiness detector. We have adopted a technique: a so-called Feature Prune Layer is placed in front of the first layer in the architecture; as a result, its weights change regarding the importance of the corresponding input features and are deleted iteratively until the desired number is reached. We test the algorithm on EEG data, as it is one of the best indicators of drowsiness based on the literature. The proposed FS algorithm is able to reduce the original feature set by 95% with only 1% degradation in precision, while the precision increases by 1.5% and 2.7% respectively when selecting the top 10% and top 20% of the initial features. Moreover, the proposed method outperforms the widely popular Principal Component Analysis and the Chi-squared test when reducing the original feature set by 95%: it achieves 24.3% and 3.2% higher precision respectively.
... Third, the results obtained from P300 detection-based devices often tend to be inaccurate because P300 detection is a "gaze-dependent process", which makes it challenging especially for people with disabilities to concentrate for an extended period [5], [79], [88], [97], [101], [107], [108], [120], [127]. Moreover, the conducted experiments are relatively unfamiliar for all subjects and have real-time constraints, so much so that once results are obtained, they cannot be changed after the recording is completed. ...
Non-invasive Visual Stimuli evoked-EEG-based P300 BCIs have gained immense attention in recent years due to their ability to help patients with disability using BCI-controlled assistive devices and applications. In addition to the medical field, P300 BCI has applications in entertainment, robotics, and education. The current article systematically reviews 147 articles that were published between 2006-2021*. Articles that pass the pre-defined criteria are included in the study. Further, classification based on their primary focus, including article orientation, participants’ age groups, tasks given, databases, the EEG devices used in the studies, classification models, and application domain, is performed. The application-based classification considers a vast horizon, including medical assessment, assistance, diagnosis, applications, robotics, entertainment, etc. The analysis highlights an increasing potential for P300 detection using visual stimuli as a prominent and legitimate research area and demonstrates a significant growth in the research interest in the field of BCI spellers utilizing P300. This expansion was largely driven by the spread of wireless EEG devices, advances in computational intelligence methods, machine learning, neural networks and deep learning.
... This EEG signals associated with the user's intentions were analyzed using several BCI paradigms. P300 [15]- [17], steadystate visual evoked potentials (SSVEP) [18], [19], and motor imagery (MI) [20]- [23] were implemented to control BCIrelated devices. Using exogenous paradigms such as SSVEP and P300 can decrease concentration and fatigue among users because they require external devices. ...
An electroencephalogram is an effective approach that provides a bidirectional pathway between the user and computer in a non-invasive way. In this study, we adopted the visual imagery data for controlling the BCI-based robotic arm. Visual imagery increases the power of the alpha frequency range of the visual cortex over time as the user performs the task. We proposed a deep learning architecture to decode the visual imagery data using only two channels and also we investigated the combination of two EEG channels that has significant classification performance. When using the proposed method, the highest classification performance using two channels in the offline experiment was 0.661. Also, the highest success rate in the online experiment using two channels (AF3-Oz) was 0.78. Our results provide the possibility of controlling the BCI-based robotic arm using visual imagery data.
... Our previous study (Li et al. 2018 Neuracle Technology (Changzhou) Co., Ltd., Beijing, China approach by flashing stimuli and scanning stimuli with the same time series. However, stimuli number increase reduces the distance between stimuli, leading to the ''flanker effect'' (Yeom and Fazli 2014) and recognition accuracy decline. This study proposes a new type of dual stimuli interface based on whole flash and local move (DS-WL), and two rules for logical grouping and presenting random. ...
Improving information transfer rate is a key to prompt the speed of outputting instructions of the event-related potential-based brain–computer interface. Our previous study designed a dual-stimuli interface that simultaneously presents two types of different stimuli to improve the speed. While, adding more stimuli into this interface makes subject easily affected by “flanker effect” that decreases the accuracy of recognizing intention. To achieve high recognition accuracy with many stimuli, this study proposes a dual stimuli interface based on whole flash and local move (DS-WL) and two rules of stimulus arrangement to induce the brain signals. Twenty subjects participated in the experiment, and their signals are recognized by a back propagation neural network classifier. The local move induces larger and later signals of targets to help discriminate the two kinds of stimuli; the rules reduce the N200 and P300 amplitudes of non-target, which improves accuracy. This study demonstrates that the DS-WL is a useful way to shorten the instruction output cycle and speed up the instructions outputting by local move and rules.
