Background: Electroencephalography (EEG)-based brain-computer interface (BCI) systems are mainly divided into three major paradigms: motor imagery (MI), event-related potential (ERP), and steady-state visually evoked potential (SSVEP). Here, we present a BCI dataset that includes the three major BCI paradigms with a large number of subjects over multiple sessions. In addition, information about the psychological and physiological conditions of BCI users was obtained using a questionnaire, and task-unrelated parameters such as resting state, artifacts, and electromyography of both arms were also recorded. We evaluated the decoding accuracies for the individual paradigms and determined performance variations across both subjects and sessions. Furthermore, we looked for more general, severe cases of BCI illiteracy than have been previously reported in the literature. Results: Average decoding accuracies across all subjects and sessions were 71.1% (± 0.15), 96.7% (± 0.05), and 95.1% (± 0.09), and rates of BCI illiteracy were 53.7%, 11.1%, and 10.2% for MI, ERP, and SSVEP, respectively. Compared to the ERP and SSVEP paradigms, the MI paradigm exhibited large performance variations between both subjects and sessions. Furthermore, we found that 27.8% (15 out of 54) of users were universally BCI literate, i.e., they were able to proficiently perform all three paradigms. Interestingly, we found no universally illiterate BCI user, i.e., all participants were able to control at least one type of BCI system. Conclusions: Our EEG dataset can be utilized for a wide range of BCI-related research questions. All methods for the data analysis in this study are supported with fully open-source scripts that can aid in every step of BCI technology. Furthermore, our results support previous but disjointed findings on the phenomenon of BCI illiteracy.

For a brain-computer interface (BCI) system, a calibration procedure is required for each individual user before he/she can use the BCI. This procedure requires approximately 20-30 min to collect enough data to build a reliable decoder. It is, therefore, an interesting topic to build a calibration-free, or subject-independent, BCI. In this article, we construct a large motor imagery (MI)-based electroencephalography (EEG) database and propose a subject-independent framework based on deep convolutional neural networks (CNNs). The database is composed of 54 subjects performing the left- and right-hand MI on two different days, resulting in 21,600 trials for the MI task. In our framework, we formulated the discriminative feature representation as a combination of the spectral-spatial input embedding the diversity of the EEG signals, as well as a feature representation learned from the CNN through a fusion technique that integrates a variety of discriminative brain signal patterns. To generate spectral-spatial inputs, we first consider the discriminative frequency bands in an information-theoretic observation model that measures the power of the features in two classes. From discriminative frequency bands, spectral-spatial inputs that include the unique characteristics of brain signal patterns are generated and then transformed into a covariance matrix as the input to the CNN. In the process of feature representations, spectral-spatial inputs are individually trained through the CNN and then combined by a concatenation fusion technique. In this article, we demonstrate that the classification accuracy of our subject-independent (or calibration-free) model outperforms that of subject-dependent models using various methods [common spatial pattern (CSP), common spatiospectral pattern (CSSP), filter bank CSP (FBCSP), and Bayesian spatio-spectral filter optimization (BSSFO)].

Brain-machine interfaces (BMIs) can be used to decode brain activity into commands to control external devices. This paper presents the decoding of intuitive upper extremity imagery for multi-directional arm reaching tasks in three-dimensional (3D) environments. We designed and implemented an experimental environment in which electroencephalogram (EEG) signals can be acquired for movement execution and imagery. Fifteen subjects participated in our experiments. We proposed a multi-directional convolution neural network-bidirectional long short-term memory network (MDCBN)-based deep learning framework. The decoding performances for six directions in 3D space were measured by the correlation coefficient (CC) and the normalized root mean square error (NRMSE) between predicted and baseline velocity profiles. The grand-averaged CCs of multi-direction were 0.47 and 0.45 for the execution and imagery sessions, respectively, across all subjects. The NRMSE values were below 0.2 for both sessions. Furthermore, in this study, the proposed MDCBN was evaluated by two online experiments for real-time robotic arm control, and the grand-averaged success rates were approximately 0.60 (±0.14) and 0.43 (±0.09), respectively. Hence, we demonstrate the feasibility of intuitive robotic arm control based on EEG signals for real-world environments.

The robust analysis of neural signals is a challenging problem. Here, we contribute a convolutional neural network (CNN) for the robust classification of a steady-state visual evoked potentials (SSVEPs) paradigm. We measure electroencephalogram (EEG)-based SSVEPs for a brain-controlled exoskeleton under ambulatory conditions in which numerous artifacts may deteriorate decoding. The proposed CNN is shown to achieve reliable performance under these challenging conditions. To validate the proposed method, we have acquired an SSVEP dataset under two conditions: 1) a static environment, in a standing position while fixated into a lower-limb exoskeleton and 2) an ambulatory environment, walking along a test course wearing the exoskeleton (here, artifacts are most challenging). The proposed CNN is compared to a standard neural network and other state-of-the-art methods for SSVEP decoding (i.e., a canonical correlation analysis (CCA)-based classifier, a multivariate synchronization index (MSI), a CCA combined with k-nearest neighbors (CCA-KNN) classifier) in an offline analysis. We found highly encouraging SSVEP decoding results for the CNN architecture, surpassing those of other methods with classification rates of 99.28% and 94.03% in the static and ambulatory conditions, respectively. A subsequent analysis inspects the representation found by the CNN at each layer and can thus contribute to a better understanding of the CNN’s robust, accurate decoding abilities.

The high feature-dimension and low sample-size problem is one of the major challenges in the study of computer-aided Alzheimer's disease (AD) diagnosis. To circumvent this problem, feature selection and subspace learning have been playing core roles in the literature. Generally, feature selection methods are preferable in clinical applications due to their ease for interpretation, but subspace learning methods can usually achieve more promising results. In this paper, we combine two different methodological approaches to discriminative feature selection in a unified framework. Specifically, we utilize two subspace learning methods, namely, linear discriminant analysis and locality preserving projection, which have proven their effectiveness in a variety of fields, to select class-discriminative and noise-resistant features. Unlike previous methods in neuroimaging studies that mostly focused on a binary classification, the proposed feature selection method is further applicable for multiclass classification in AD diagnosis. Extensive experiments on the Alzheimer's disease neuroimaging initiative dataset showed the effectiveness of the proposed method over other state-of-the-art methods.