EEG and Brain-Computer Interfaces

Generalized neural decoders for transfer learning across participants and recording modalities

This paper introduces HTNet, a neural decoder that generalizes across participants and recording modalities, achieving performance comparable to tailored decoders with minimal data.

Stimulation Artifact Source Separation (SASS) for assessing electric brain oscillations during transcranial alternating current stimulation (tACS)

This paper introduces SASS, an algorithm to remove stimulation artifacts during AM-tACS, enabling reliable assessment of brain oscillations and enabling adaptive stimulation.

Steady-state visually evoked potentials and feature-based attention: Pre-registered null results and a focused review of methodological considerations.

This study reports null results for attention-related modulation of SSVEP responses to color cues and reviews methodological considerations for SSVEP studies of feature-based attention.

Selective modulation of population dynamics during neuroprosthetic skill learning

This study examines how neural population dynamics change in direct and indirect subpopulations during BMI learning, finding that coordination increases mainly in the direct subpopulation.

In search for the most optimal EEG method: A practical evaluation of a water-based electrode EEG system.

The paper evaluates a water-based electrode EEG system compared to a gel-based system, finding higher noise and data loss in the water-based system affecting signal quality and analysis.

A BCI-based vibrotactile neurofeedback training improves motor cortical excitability during motor imagery

This study shows that vibrotactile feedback during motor imagery BCI training increases motor cortical excitability and mu-rhythm desynchronization.

Typical somatomotor physiology of the hand is preserved in a patient with an amputated arm

The paper investigates motor cortex electrophysiology in an amputee, finding preserved hand representations and high decoding accuracy for finger movements.

ScoreNet: A Neural network-based post-processing model for identifying epileptic seizure onset and offset in EEGs

ScoreNet improves seizure onset and offset detection in EEGs by using a neural network with a custom loss function, achieving higher F1 scores and lower false positive rates.

The neural representation of force across grasp types in motor cortex of humans with tetraplegia

This study examines how force and grasp are represented in motor cortex of tetraplegic individuals, finding that grasp affects force decoding and that grasp is more accurately classified than force.

Prediction of Inefficient BCI Users based on Cognitive Skills and Personality Traits

This study predicts BCI inefficiency using cognitive skills and personality traits, finding that visuospatial memory predicts suppression during motor imagery tasks.

A group refactorization procedure for sleep electroencephalography

The paper introduces a group refactorization method for sleep EEG data to enable inter-individual analysis by aligning frequency factors across subjects.

Structural templates for imaging EEG cortical sources in infants

This paper introduces 12 anatomical templates for EEG source reconstruction in infants aged 0-24 months, validated using EEG recordings and MRI data.

Objective Extraction of Evoked Event-related Oscillations from Time-frequency Representation of Event-related Potentials

This paper introduces a data-driven method to objectively extract evoked event-related oscillations from ERP data, showing improved detection compared to conventional methods.

Signal-space projection suppresses the tACS artifact in EEG recordings

This study evaluates artifact correction techniques for tACS-contaminated EEG data, finding that signal-space projection (SSP) effectively suppresses the tACS artifact while preserving brain activity.

Sensorimotor functional connectivity: a neurophysiological factor related to BCI performance

This study identifies sensorimotor functional connectivity as a neurophysiological factor correlated with BCI performance, showing its significance beyond oscillation amplitude.

DN3: An open-source Python library for large-scale raw neurophysiology data assimilation for more flexible and standardized deep learning

DN3 is an open-source Python library for deep learning with neurophysiology data, enabling reproducible analysis and benchmarking across datasets, with applications in EEG-based biometrics.

Mapping Cognitive Brain Functions at Scale

This paper introduces a mobile platform called BrainE for rapid, scalable EEG-based assessment of multiple cognitive tasks and demonstrates its reliability and sensitivity to neural plasticity and mental health symptoms.

Motor Imagery EEG Neurofeedback Skill Acquisition in the Context of Declarative Interference and Sleep

This study investigates whether sleep and declarative interference affect motor imagery neurofeedback skill acquisition, finding no significant effects on ERD.

Shared Component Analysis

This paper introduces Shared Component Analysis (SCA) as an alternative to PCA for neuroimaging data dimensionality reduction, prioritizing components shared across multiple sensors.

Are Sources of EEG and MEG rhythmic activity the same? An analysis based on BC-VARETA

The paper compares EEG and MEG source spectra using BC-VARETA and finds no differences in reconstructed sources between the two modalities.

A P300 brain-computer interface with a reduced visual field

This paper investigates reducing the visual field in a P300 speller to improve performance by minimizing non-target stimulus contributions.

An Evolving-Dynamic Network Activity Approach toEpileptic Seizure Prediction using Machine Learning

This paper uses machine learning models to predict epileptic seizures in mice, achieving 98% accuracy with logistic regression and convolutional neural networks.

Early Classification of Motor Tasks Using Dynamic Functional Connectivity Graphs from EEG

This paper introduces a method to classify motor tasks from EEG signals with high accuracy using short recording intervals, achieving 85.32% accuracy within 500 ms.

Impact of referencing scheme on decoding performance of LFP-based brain-machine interface

The paper evaluates how different referencing schemes affect LFP-based BMI decoding performance and finds that CAR with LMP features improves long-term decoding.

Detection of Static, Dynamic, and No Tactile Friction Based on Non-linear dynamics of EEG Signals: A Preliminary Study

This study investigates how different types of tactile friction affect EEG non-linear features and finds that non-linear indices can distinguish between static, dynamic, and no friction states.

Decoding spoken English phonemes from intracortical electrode arrays in dorsal precentral gyrus

This study decodes 39 English phonemes from neural signals in the dorsal precentral gyrus using intracortical electrode arrays, achieving 33.9% classification accuracy with a recurrent neural network.

Short-Delay Neurofeedback Facilitates Training of the Parietal Alpha Rhythm

This study shows that short-latency visual neurofeedback of parietal alpha activity enhances training effectiveness, with longer delays reducing sustained changes in alpha rhythm.

Unsupervised Learning of Brain State Dynamics during Emotion Imagery using High-Density EEG

This study uses unsupervised AMICA decomposition to analyze high-density EEG data during emotion imagery, identifying distinct brain state dynamics across different emotions.

Fast EEG-based decoding of the directional focus of auditory attention using common spatial patterns

This paper introduces a method using filterbank common spatial patterns to decode auditory attention direction from EEG, achieving high accuracy on short signal segments and demonstrating adaptability to new subjects and limited channel setups.

Determinants of Real-Time fMRI Neurofeedback Performance and Improvement: a Machine Learning Mega-Analysis

This study uses machine learning to analyze 608 participants and identifies two factors improving neurofeedback performance: pre-training no-feedback runs and training patients over healthy participants.