Functional near-infrared spectroscopy (fNIRS) is used to measure cerebral activity because it simple and portable. However, scalp-hemodynamics often contaminates fNIRS signals, leading detection of cortical in regions that are actually inactive. Methods for removing these artifacts using standard source–detector distance channels (Long-channel) tend over-estimate the artifacts, while methods additional short (Short-channel) require numerous probes cover broad areas, which leads a high cost...

Many researchers have used machine learning models to control artificial hands, walking aids, assistance suits, etc., using the biological signal of electromyography (EMG). The use such devices requires high classification accuracy. One method for improving performance is normalization, as z-score. However, normalization not in most EMG-based motion prediction studies because need calibration and fluctuation reference value (cannot re-use). Therefore, this study, we proposed a that combines...

Recently, Auditory Attention Decoding (AAD), a method that uses information on the listener's auditory intention (electroencephalogram) to identify attended speech stream among multiple speech, has been attracting attention. This can enhance only and reduce unattended stream. Although previous research demonstrated high decoding accuracy in limited experiments for single sound, practical use, it is necessary assume an environment closer real life. Therefore, this study, experiment was...

Performing a complex sequential finger movement requires the temporally well-ordered organization of individual movements. Previous behavioural studies have suggested that brain prepares whole sequence movements as single set, rather than fingers. However, direct neuroimaging support for this hypothesis is lacking and, assuming it to be true, remains unclear which regions represent information prepared sequence. Here, we measured activity with functional magnetic resonance imaging while 14...

Humans achieve efficient behaviors by perceiving and responding to errors. Error-related potentials (ErrPs) are electrophysiological responses that occur upon Leveraging ErrPs improve the accuracy of brain-computer interfaces (BCIs), utilizing brain's natural error-detection processes enhance system performance, has been proposed. However, influence external contextual factors on detectability remains poorly understood, especially in multitasking scenarios involving both BCI operations...

Brain-machine interfaces (BMI) rely on the accurate classification of event-related potentials (ERPs) and their performance greatly depends appropriate selection classifier parameters features from dense-array electroencephalography (EEG) signals. Moreover, in order to achieve a portable more compact BMI for practical applications, it is also desirable use system capable using information as few EEG channels possible. In present work, we propose method classifying P300 ERPs combination...

Functional near-infrared spectroscopy (fNIRS) is expected to be applied brain-computer interface (BCI) technologies. Since lengthy fNIRS measurements are uncomfortable for participants, it difficult obtain enough data train classification models; hence, the fNIRS-BCI accuracy decreases.In this study, improve accuracy, we examined an augmentation method using Wasserstein generative adversarial networks (WGANs). Using during hand-grasping tasks, evaluated whether proposed could generate...

The auditory Brain-Computer Interface (BCI) using electroencephalograms (EEG) is a subject of intensive study. As cue, BCIs can deal with many the characteristics stimuli such as tone, pitch, and voices. Spatial information on also provides useful for BCI. However, in portable system, virtual have to be presented spatially through earphones or headphones, instead loudspeakers. We investigated possibility an BCI out-of-head sound localization technique, which enables us present users from any...

Motor action is prepared in the human brain for rapid initiation at appropriate time. Recent non-invasive decoding techniques have shown that activity preparation represents various parameters of an upcoming action. In present study, we demonstrated a freely chosen effector can be predicted from measured using functional magnetic resonance imaging (fMRI) before Furthermore, was related to response time (RT). We with fMRI while 12 participants performed finger-tapping task either left or...

Abstract Why does Fitts’ law fit various human behavioural data well even though it is not a model based on physical dynamics? To clarify this, we derived the relationships among factors applied in law—movement duration and spatial endpoint error—based multi-joint forward- inverse-dynamics models presence of signal-dependent noise. As result, relationship between them was modelled as an inverse proportion. validate whether error calculated by can represent actual movements, conducted...

From allowing basic communication to move through an environment, several attempts are being made in the field of brain-computer interfaces (BCI) assist people that somehow find it difficult or impossible perform certain activities. Focusing on these as potential users BCI, we obtained electroencephalogram (EEG) readings from nine healthy subjects who were presented with auditory stimuli via earphones six different virtual directions. We following oddball paradigm elicit P300 waves within...

Recently, a brain-computer interface (BCI) using virtual sound sources has been proposed for estimating user intention via electroencephalogram (EEG) in an oddball task. However, its performance is still insufficient practical use. In this study, we examine the impact that shortening stimulus onset asynchrony (SOA) on auditory BCI. While very short SOA might improve performance, perception and task become difficult, event-related potentials (ERPs) may not be induced if too short. Therefore,...

Recent functional magnetic resonance imaging (fMRI) decoding techniques allow us to predict the contents of sensory and motor events or participants' mental states from multi-voxel patterns fMRI signals. Sparse logistic regression (SLR) is a useful pattern classification algorithm that has advantage being able automatically select voxels avoid over-fitting. However, SLR suffers over-pruning, in which many are potentially for prediction discarded.We propose an ensemble solution called...

In the study of brain computer interface using electroencephalogram (EEG), area motor imagery individual finger movement has been extensively examined. The objective this is to predict which moving or being imaged through use EEG. We measured EEG activity while subjects performed either execution movements with their right hand. Event related spectral perturbation was used as indicator activity, represents frequency power fluctuation from baseline interval. bands α, β, and γ (8-15, 16-31,...

As brain-computer interfaces (BCI) must provide reliable ways for end users to accomplish a specific task, methods secure the best possible translation of intention are constantly being explored. In this paper, we propose and test number convolutional neural network (CNN) structures identify classify single-trial P300 in electroencephalogram (EEG) readings an auditory BCI. The recorded data correspond nine subjects series experiment sessions which stimuli following oddball paradigm were...

Procedural motor learning includes a period when no substantial gain in performance improvement is obtained even with repeated, daily practice. Prompted by the potential benefit of high-frequency transcutaneous electrical stimulation, we examined if stimulation to hand reduces redundant activity that likely exists an acquired skill, so as further upgrade stable performance. Healthy participants were trained until their continuously rotating two balls palm right became stable. In series...

Once people have a well-trained motor skill, their performance becomes stabilized and achieving substantial improvement is difficult. Recently, we shown that even plateaued hand skill can be upgraded with short-period electrical stimulation to the prior task. Here, identify neuronal substrates underlying of by examining enhanced functional connectivity in sensory-motor regions are associated learning. We measured brain activity using magnetic resonance imaging performed psychophysiological...

The implementation of a brain–computer interface (BCI) using electroencephalography typically entails two phases: feature extraction and classification utilizing classifier. Consequently, there are numerous disordered combinations techniques that apply to each target dataset. In this study, we employed neural network as classifier address the versatility system in converting inputs various forms into outputs forms. As preprocessing step, utilized transposed convolution augment width number...

Herein, we investigated the effects of using time segments, including visual presentation, motor imagery, and rest time, as training data in a brain-computer interface (BCI) competition. Using BCI Competition IV 2a 2b, many researchers have attempted to create more robust classifiers with higher classification accuracy. Some studies also used presentation data. However, use outside imagery makes comparisons performance across models difficult, may lead that are overfitted experimental...

Functional near-infrared spectroscopy (fNIRS) is a widely utilized neuroimaging tool in fundamental neuroscience research and clinical investigation. Previous has revealed that task-evoked systemic artifacts mainly originating from the superficial-tissue may preclude identification of cerebral activation during given task. We examined influence such on event-related brain activity brisk squeezing movement. estimated hemodynamics short source–detector distance channels (15 mm) by applying...

Functional near-infrared spectroscopy (fNIRS) is expected to be applied the brain-computer interface (BCI). Since a lengthy fNIRS measurement uncomfortable for participant, it difficult obtain sufficient amount of data train classification models; hence, fNIRS-BCI accuracy decreases. In this study, improve accuracy, we examined an data-augmentation method using generative adversarial networks (GANs). Using simulation data, evaluated whether proposed could generate artificial data. Comparing...

Functional near-infrared spectroscopy (fNIRS) is an effective non-invasive neuroimaging technique for measuring hemoglobin concentration in the cerebral cortex. Owing to nature of fNIRS measurement principles, measured signals can be contaminated with task-related scalp blood flow (SBF), which distributed over whole head and masks true brain activity. Aiming fNIRS-based real-time application, we proposed a SBF artifact reduction method. Using principal component analysis, estimated global...

Measuring discrete-trial motor-related brain activity using functional near-infrared spectroscopy (fNIRS) is considered difficult. This because its spatial resolution much lower than that of magnetic resonance imaging (fMRI), and signals include non-motion-related artifacts. To detect changes in hemoglobin induced by movements, most fNIRS studies have used a block design which subject conducts set repetitive movements for over few seconds. Changes the series are accumulated. Here, we address...