EEG and Brain-Computer Interfaces

MEEGNet: An open source python library for the application of convolutional neural networks to MEG

MEEGNet is a Python library for applying convolutional neural networks to MEG data, achieving high accuracy in classification tasks while offering interpretability tools.

Spatial (Mis)match Between EEG and fMRI Signal Patterns Revealed by Spatio-Spectral Source-Space EEG Decomposition

This paper compares EEG and fMRI data using spatio-spectral decomposition, finding that EEG patterns show weak but significant spatial similarity to fMRI BOLD signatures, with complementary information.

Classification of motor-related brain states using high frequency information from muscle recordings

The paper classifies motor-related brain states using EMG and EEG frequency bands, finding comparable accuracy in decoding cancellation states from both modalities.

Deepening sleep using an EEG wearable featuring modeling-based closed-loop neurostimulation

This paper demonstrates that modeling-based closed-loop neurostimulation using an EEG headband can enhance slow-wave sleep through acoustic stimulation of slow oscillations.

Less is more: selection from a small set of options improves BCI velocity control

This paper introduces a discrete direction selection decoder for iBCI that outperforms continuous velocity decoders in human and monkey experiments by improving control accuracy and reducing noise.

Automatic detection of nociceptive pain levels using frequency bands from electroencephalographic (EEG) signals

This paper uses EEG signals to detect nociceptive pain levels, achieving 83% accuracy in distinguishing high and low pain based on frequency band power.

Enhanced human sensorimotor integration via self-modulation of the somatosensory activity

This study shows that closed-loop neurofeedback training during motor imagery enhances somatosensory cortex activity and improves keyboard typing performance.

Enhanced EEG Forecasting: A Probabilistic Deep Learning Approach

This paper applies probabilistic deep learning to forecast resting-state EEG signals, achieving more accurate long-term predictions than autoregressive models.

Flexible regulation of representations on a drifting manifold enables long-term stable complex neuroprosthetic control

This study shows that neural representations for imagined movements remain stable over time, allowing long-term neuroprosthetic control without recalibration.

Learning to operate an imagined speech Brain-Computer Interface involves the spatial and frequency tuning of neural activity

This study shows that training improves BCI control through changes in EEG power and neural features, suggesting the need for both machine and human learning in BCI development.

Block-Champagne: A novel Bayesian framework for Imaging E/MEG Source

Block-Champagne improves E/MEG source reconstruction by accurately estimating both source locations and extents using a Bayesian framework with block-sparsity constraints and multimodal priors.

Utilizing Pretrained Vision Transformers and LargeLanguage Models for Epileptic Seizure Prediction

This paper uses pre-trained Vision Transformers and Large Language Models to predict epileptic seizures, finding that LLMs achieve higher sensitivity than ViTs and a custom ResNet model.

GREEN: a lightweight architecture using learnable wavelets and Riemannian geometry for biomarker exploration

GREEN is a lightweight neural network that combines wavelet transforms and Riemannian geometry for EEG biomarker analysis, outperforming non-deep models and large DL models with fewer parameters.

Real-time control of a hearing instrument with EEG-based attention decoding

This paper presents a real-time BCI system using EEG attention decoding to control a hearing instrument, enhancing speech perception in noisy environments by steering attention to desired sound sources.

Decoding in the fourth dimension: Classification of temporal patterns and their generalization across locations

The paper introduces a decoding method using temporal features from neural time series to classify experimental conditions and assess generalization across sensor locations.

How different immersive environments affect intracortical brain computer interfaces

This study examines how immersive environments affect BCI performance, finding that prior experience with an environment is more important than its immersiveness for decoder performance and generalization.

Improving wearable-based seizure prediction by feature fusion using growing network

This paper improves seizure prediction using a growing neural network that fuses features from LSTMs, DCCAEs, and time-of-day patterns, achieving higher accuracy than baseline methods.

Unsupervised, piecewise linear decoding enables an accurate prediction of muscle activity in a multi-task brain computer interface

This paper develops an unsupervised piecewise linear decoder for multi-task brain-computer interfaces, showing it outperforms global decoders in EMG prediction.

A Brain-Computer Interface for Improving Auditory Attention in Multi-Talker Environments

This paper develops a neurofeedback system using EEG to improve auditory attention in multi-talker environments by attenuating unattended talkers based on real-time attention decoding.

Behavioral Classification of Sequential Neural Activity Using Time Varying Recurrent Neural Networks

This paper introduces Time-Varying RNNs (TV-RNNs) for early classification of neural activity sequences, showing improved accuracy and earlier detection compared to standard RNNs in mouse and human motor task data.

Combining video telemetry and wearable MEG for naturalistic imaging

This paper combines video telemetry with wearable MEG to analyze naturalistic movement and neural activity during dance.

Brain responses vary in duration - modeling strategies and challenges

The paper shows that ignoring event durations in brain signal analysis can lead to spurious results and proposes non-linear spline regression as a better method for modeling duration effects, compatible with overlap correction.

Blind source separation of event-related potentials using a recurrent neural network

This paper develops a recurrent neural network for blind source separation of ERP difference waveforms, showing improved separation compared to ICA.

Exploring dynamic brain oscillations in motor imagery and low-frequency sound

This study examines brain activity during motor imagery and low-frequency sound listening, finding variability in alpha and beta power responses but no additional modulation from combining the two tasks.

Fast and Accurate EEG/MEG BEM-Based Forward Problem Solution for High-Resolution Head Models

The paper develops a BEM-based method to solve EEG/MEG forward problems for high-resolution head models in about 60 seconds.

Localization of Realistic Spatial Patches of Complex Source Activity in MEG

The paper introduces PATCH-AP, a source localization method for MEG that outperforms existing techniques in localizing both discrete and extended neural sources.

Acoustic inspired brain-to-sentence decoder for logosyllabic language

This paper develops a brain-to-sentence decoder for Mandarin Chinese, achieving 71% character accuracy by leveraging acoustic features and language models.

SleepSatelightFTC: A Lightweight and Interpretable Deep Learning Model for Single-Channel EEG-Based Sleep Stage Classification

The paper introduces SleepSatelightFTC, a lightweight deep learning model for EEG-based sleep stage classification that achieves high accuracy with fewer parameters and provides interpretability through self-attention.

Electroencephalogram (EEG) Classification using a bio-inspired Deep Oscillatory Neural Network

This paper introduces two oscillatory neural networks for EEG classification and prediction tasks, incorporating brain dynamics through nonlinear oscillators.

Preprocessing Choices for P3 Analyses with Mobile EEG: A Systematic Literature Review and Interactive Exploration

This paper reviews preprocessing choices for P3 ERP analysis in mobile EEG, finds inconsistent reporting, and provides an interactive tool for exploring and comparing preprocessing strategies.