EEG and Brain-Computer Interfaces

Neural population variance explains adaptation differences during learning

The paper investigates how neural spiking variability affects adaptation during a brain-computer interface task in rhesus macaques, finding that population variance metrics differ between easy and hard adaptation conditions.

RELAX-Jr: An Automated Pre-Processing Pipeline for Developmental EEG Recordings

RELAX-Jr is an automated EEG pre-processing pipeline designed for children's data, showing strong cleaning performance across multiple metrics compared to existing pipelines.

Robust Detection of Brain Stimulation Artifacts in iEEG Using Autoencoder-Generated Signals and ResNet Classification

This paper develops a method to detect brain stimulation artifacts in iEEG using synthetic data and a ResNet classifier, achieving high accuracy on real data.

Introducing Haptic Feedback in Real-Time fMRI Neurofeedback: A Novel Approach to Modulate Primary Motor Cortex Activity

This study compares visuo-haptic feedback with visual and haptic feedback in fMRI neurofeedback for motor imagery tasks, finding that combined feedback better activates the primary motor cortex.

Representational dissimilarity component analysis (ReDisCA)

ReDisCA estimates spatial-temporal components in EEG/MEG data aligned with a target RDM, improving source localization without inverse modeling.

Towards EEG-to-Text: Handwritten Character Classification via Continuous Kinematic Decoding

This study classifies handwritten letters from non-invasive EEG signals using direct classification and a two-step kinematic decoding approach, showing improved performance with the latter.

Influence of feedback transparency on motor imagery neurofeedback performance: the contribution of agency

This study examines how feedback transparency in neurofeedback affects performance and agency, finding that transparent feedback improves performance while motor illusions reduce it.

EEG Electrodes and Where to Find Them: Automated Localization From 3D Scans

This paper presents a method for accurately localizing EEG electrode positions using 3D scans, validated on a large dataset, achieving 85.7% to 91.0% accuracy without requiring MRI.

Sensorimotor brain-computer interface performance depends on signal-to-noise ratio but not connectivity of the mu rhythm in a multiverse analysis of longitudinal data

This study investigates BCI performance using longitudinal data and finds that signal-to-noise ratio, not mu rhythm connectivity, predicts performance across multiple analysis pipelines.

Localized estimation of event-related neural source activity from simultaneous MEG-EEG with a recurrent neural network

This paper introduces a recurrent neural network for estimating neural sources from MEG-EEG data, showing improved performance over traditional methods.

Diffusion model-based image generation from rat brain activity

This paper generates images from rat brain activity using a latent diffusion model, converting continuous neural signals into visual outputs.

Evaluating noise correction approaches for non-invasive electrophysiology of the human spinal cord

This study evaluates noise correction methods for electrospinography (ESG) to reduce cardiac artefacts and improve signal quality in human spinal cord recordings.

Cycle-frequency content EEG analysis improves the assessment of respiratory-related cortical activity

This paper introduces Cycle-Frequency (C-F) analysis to improve the assessment of respiratory-related cortical activity by enhancing time and frequency localization compared to conventional Time-Frequency analysis.

Distraction Impact of Concurrent Conversation on Event-Related Potential Based Brain-Computer Interfaces

This study examines how concurrent conversation affects ERP-based BCIs, finding that speaking degrades accuracy and ERP amplitude while listening has less impact.

A comparative study between a power and a connectivity sEEG biomarker for seizure-onset zone identification in temporal lobe epilepsy

This study compares power and connectivity-based sEEG biomarkers for identifying seizure-onset zones in temporal lobe epilepsy, finding that power-based methods generally perform better but both can achieve similar results with optimal parameter tuning.

Automatic detection of seizures within zebrafish larvae epilepsy models using single-channel electroencephalography

This paper develops and evaluates seizure detection algorithms for zebrafish EEG data to improve efficiency in epilepsy research.

EEG source imaging technique to investigate sleep oscillations for simultaneous EEG-fMRI

This paper applies beamforming spatial filtering to EEG-fMRI data to improve detection of sleep spindles and enhance fMRI GLM analysis by reducing BCG artefacts.

Convolutional neural networks can identify brain interactions involved in decoding spatial auditory attention

This study uses CNNs to identify brain interactions involved in spatial auditory attention from EEG data, achieving high classification accuracy and revealing specific frequency-band interactions.

EEG Analyses of visual cue effects on executed movements

The paper evaluates four visual cues to determine their impact on brain signals during movement tasks, finding that Rotation and Reference cues minimize variability and improve classifier accuracy for brain-computer interfaces.

Image retrieval based on closed-loop visual-semantic neural decoding

This paper demonstrates that closed-loop brain-computer interfaces can retrieve intended images using a CLIP latent space, achieving over 80% accuracy, while failing with AlexNet.

The influence of the head model on magnetoencephalography-derived functional connectivity fingerprinting

This study shows that the head model significantly affects the accuracy of subject identification and differentiation in MEG-derived functional connectivity fingerprinting.

Machine learning of brain-specific biomarkers from EEG

This study uses machine learning to analyze EEG data and shows that both brain and peripheral signals contribute to predicting age and sex, but brain signals are more effective for CNS-specific biomarkers.

Neural subspaces of imagined movements in parietal cortex remain stable over several years in humans.

This study shows that neural subspaces for imagined movements in the parietal cortex remain stable over several years in humans, improving the long-term reliability of brain-machine interfaces.

Brain-responsive music enables non-invasive, targeted and unobtrusive neurostimulation

This paper demonstrates brain-responsive music can modulate neural oscillations by using real-time brain signal feedback to control musical elements, enabling targeted neurostimulation.

MLSPred-Bench: ML-Ready Benchmark LeveragingSeizure Detection EEG data for Predictive Models

The paper converts seizure detection EEG data into ML-ready datasets for predictive modeling, creating MLSPred-Bench with 12 datasets and 150GB of data.

Leveraging Grating-Based Flickers: A Leap Toward Practical, Visually Comfortable, and High-Performance Dry EEG Code-VEP BCI

This paper introduces StAR stimuli for dry EEG Code-VEP BCI, showing improved accuracy and visual comfort compared to traditional flickers.

Laying the Foundation: Modern Transformers for Gold-Standard Sleep Analysis and Beyond

The paper evaluates transformer-based models for sleep stage classification, finding they match state-of-the-art techniques and provide a foundation for future data-driven sleep research.

Self-Supervised Transformer Model Training for a Sleep-EEG Foundation Model

This paper introduces a self-supervised transformer model for sleep-EEG data that outperforms supervised methods and can be fine-tuned for new tasks like individual identification and brain age estimation.

Detection of High-Frequency Oscillations from Intracranial EEG Data with Switching State Space Model

This paper uses a Switching State Space Model to automatically detect high-frequency oscillations in intracranial EEG data, improving accuracy compared to manual annotation.

Evaluating Augmentation Approaches for Deep Learning-based Major Depressive Disorder Diagnosis with Raw Electroencephalogram Data

This study evaluates six EEG data augmentation methods for MDD diagnosis, finding that most do not improve performance beyond using duplicate data, but channel dropout augmentation does.