Toward Domain-Free Transformer for Generalized EEG Pre-Training.

Electroencephalography (EEG) signals are the brain signals acquired using the non-invasive approach. Owing to the high portability and practicality, EEG signals have found extensive application in monitoring human physiological states across various domains. In recent years, deep learning methodologies have been explored to decode the intricate information embedded in EEG signals. However, since EEG signals are acquired from humans, it has issues with acquiring enormous amounts of data for training the deep learning models. Therefore, previous research has attempted to develop pre-trained models that could show significant performance improvement through fine-tuning when data are scarce. Nonetheless, existing pre-trained models often struggle with constraints, such as the necessity to operate within datasets of identical configurations or the need to distort the original data to apply the pre-trained model. In this paper, we proposed the domain-free transformer, called DFformer, for generalizing the EEG pre-trained model. In addition, we presented the pre-trained model based on DFformer, which is capable of seamless integration across diverse datasets without necessitating architectural modification or data distortion. The proposed model achieved competitive performance across motor imagery and sleep stage classification datasets. Notably, even when fine-tuned on datasets distinct from the pre-training phase, DFformer demonstrated marked performance enhancements. Hence, we demonstrate the potential of DFformer to overcome the conventional limitations in pre-trained model development, offering robust applicability across a spectrum of domains.

Decoding Neural Correlation of Language-Specific Imagined Speech using EEG Signals.

Speech impairments due to cerebral lesions and degenerative disorders can be devastating. For humans with severe speech deficits, imagined speech in the brain-computer interface has been a promising hope for reconstructing the neural signals of speech production. However, studies in the EEG-based imagined speech domain still have some limitations due to high variability in spatial and temporal information and low signal-to-noise ratio. In this paper, we investigated the neural signals for two groups of native speakers with two tasks with different languages, English and Chinese. Our assumption was that English, a non-tonal and phonogram-based language, would have spectral differences in neural computation compared to Chinese, a tonal and ideogram-based language. The results showed the significant difference in the relative power spectral density between English and Chinese in specific frequency band groups. Also, the spatial evaluation of Chinese native speakers in the theta band was distinctive during the imagination task. Hence, this paper would suggest the key spectral and spatial information of word imagination with specialized language while decoding the neural signals of speech. Clinical Relevance- Imagined speech-related studies lead to the development of assistive communication technology especially for patients with speech disorders such as aphasia due to brain damage. This study suggests significant spectral features by analyzing cross-language differences of EEG-based imagined speech using two widely used languages.

Noise-induced hearing loss in zebrafish model: Characterization of tonotopy and sex-based differences.

Hearing loss caused by frequent and persistent exposure to loud noise is one of the most common diseases in modern society. Many studies have demonstrated the characteristics of noise-induced hearing loss in human and non-human vertebrate models, including frequency-specific noise-induced hearing loss and sex-biased differences. Zebrafish (Danio rerio) is a useful hearing research model because its lateral line is easy to access and because of its detailed perception of sound. Despite the increasing popularity of zebrafish as a model for NIHL, a better understanding of this model is needed to determine sex differences in NIHL. To study the features of zebrafish as they relate to an NIHL model, we tested various phenotypes after frequency-specific noise stimulation. The degree of damage to hair cells and hearing loss were investigated after exposing zebrafish to 200 Hz and 1 kHz continuous waves and broadband white noise with a bandwidth from 50 Hz to 1 kHz. After exposure to all frequencies, the larvae showed lateral line hair cell damage, which is superficially located. In adult zebrafish, the threshold of auditory-evoked potential signals is elevated. Moreover, the number of hair cells remarkably decreased in the rostral region of the saccule, after exposure to 1 kHz and white noise, whereas zebrafish exposed to 200 Hz noise showed a decrease in hair cells in the caudal region. Moreover, male zebrafish were found to be more vulnerable to noise than female zebrafish, as is the case in humans and other mammals. Cortisol levels also increased in the noise-exposed male group, as compared to the noise-exposed female and control male groups. However, there was no difference in cortisol levels when the noise-exposed female group was compared to the control female group. Our study demonstrates not only that noise-induced hearing loss is frequency-dependent but also that the degree of hearing loss is affected by sex in zebrafish, emphasizing the need to consider sex in NIHL studies.

Multiscale Convolutional Transformer for EEG Classification of Mental Imagery in Different Modalities.

A new kind of sequence-to-sequence model called a transformer has been applied to electroencephalogram (EEG) systems. However, the majority of EEG-based transformer models have applied attention mechanisms to the temporal domain, while the connectivity between brain regions and the relationship between different frequencies have been neglected. In addition, many related studies on imagery-based brain-computer interface (BCI) have been limited to classifying EEG signals within one type of imagery. Therefore, it is important to develop a general model to learn various types of neural representations. In this study, we designed an experimental paradigm based on motor imagery, visual imagery, and speech imagery tasks to interpret the neural representations during mental imagery in different modalities. We conducted EEG source localization to investigate the brain networks. In addition, we propose the multiscale convolutional transformer for decoding mental imagery, which applies multi-head attention over the spatial, spectral, and temporal domains. The proposed network shows promising performance with 0.62, 0.70, and 0.72 mental imagery accuracy with the private EEG dataset, BCI competition IV 2a dataset, and Arizona State University dataset, respectively, as compared to the conventional deep learning models. Hence, we believe that it will contribute significantly to overcoming the limited number of classes and low classification performances in the BCI system.

Multimodal signal dataset for 11 intuitive movement tasks from single upper extremity during multiple recording sessions.

BACKGROUND: Non-invasive brain-computer interfaces (BCIs) have been developed for realizing natural bi-directional interaction between users and external robotic systems. However, the communication between users and BCI systems through artificial matching is a critical issue. Recently, BCIs have been developed to adopt intuitive decoding, which is the key to solving several problems such as a small number of classes and manually matching BCI commands with device control. Unfortunately, the advances in this area have been slow owing to the lack of large and uniform datasets. This study provides a large intuitive dataset for 11 different upper extremity movement tasks obtained during multiple recording sessions. The dataset includes 60-channel electroencephalography, 7-channel electromyography, and 4-channel electro-oculography of 25 healthy participants collected over 3-day sessions for a total of 82,500 trials across all the participants. FINDINGS: We validated our dataset via neurophysiological analysis. We observed clear sensorimotor de-/activation and spatial distribution related to real-movement and motor imagery, respectively. Furthermore, we demonstrated the consistency of the dataset by evaluating the classification performance of each session using a baseline machine learning method. CONCLUSIONS: The dataset includes the data of multiple recording sessions, various classes within the single upper extremity, and multimodal signals. This work can be used to (i) compare the brain activities associated with real movement and imagination, (ii) improve the decoding performance, and (iii) analyze the differences among recording sessions. Hence, this study, as a Data Note, has focused on collecting data required for further advances in the BCI technology.

Classification of Drowsiness Levels Based on a Deep Spatio-Temporal Convolutional Bidirectional LSTM Network Using Electroencephalography Signals.

Non-invasive brain-computer interfaces (BCI) have been developed for recognizing human mental states with high accuracy and for decoding various types of mental conditions. In particular, accurately decoding a pilot's mental state is a critical issue as more than 70% of aviation accidents are caused by human factors, such as fatigue or drowsiness. In this study, we report the classification of not only two mental states (i.e., alert and drowsy states) but also five drowsiness levels from electroencephalogram (EEG) signals. To the best of our knowledge, this approach is the first to classify drowsiness levels in detail using only EEG signals. We acquired EEG data from ten pilots in a simulated night flight environment. For accurate detection, we proposed a deep spatio-temporal convolutional bidirectional long short-term memory network (DSTCLN) model. We evaluated the classification performance using Karolinska sleepiness scale (KSS) values for two mental states and five drowsiness levels. The grand-averaged classification accuracies were 0.87 (±0.01) and 0.69 (±0.02), respectively. Hence, we demonstrated the feasibility of classifying five drowsiness levels with high accuracy using deep learning.

Engineering Titanium Dioxide Nanostructures for Enhanced Lithium-Ion Storage.

Various kinds of nanostructured materials have been extensively investigated as lithium ion battery electrode materials derived from their numerous advantageous features including enhanced energy and power density and cyclability. However, little is known about the microscopic origin of how nanostructures can enhance lithium storage performance. Herein, we identify the microscopic origin of enhanced lithium storage in anatase TiO2 nanostructure and report a reversible and stable route to achieve enhanced lithium storage capacity in anatase TiO2. We designed hollow anatase TiO2 nanostructures composed of interconnected ∼5 nm sized nanocrystals, which can individually reach the theoretical lithium storage limit and maintain a stable capacity during prolonged cycling (i.e., 330 mAh g-1 for the initial cycle and 228 mAh g-1 for the 100th cycle, at 0.1 A g-1). In situ characterization by X-ray diffraction and X-ray absorption spectroscopy shows that enhanced lithium storage into the anatase TiO2 nanocrystal results from the insertion reaction, which expands the crystal lattice during the sequential phase transition (anatase TiO2 → Li0.55TiO2 → LiTiO2). In addition to the pseudocapacitive charge storage of nanostructures, our approach extends the utilization of nanostructured TiO2 for significantly stabilizing excess lithium storage in crystal structures for long-term cycling, which can be readily applied to other lithium storage materials.