EEG and Brain-Computer Interfaces

Across-subject offline decoding of motor imagery from MEG and EEG

This study compares methods for inter-subject decoding of motor imagery from MEG and EEG, finding that multi-task learning with logistic regression and l2,1-norm regularization achieves the best accuracy when trained on motor imagery data.

Data-driven re-referencing of intracranial EEG based on independent component analysis (ICA)

This paper challenges bipolar referencing in intracranial EEG and proposes ICA as a more effective method for isolating local brain activity.

Populational Super-Resolution Sparse M/EEG Sources and Connectivity Estimation

This paper introduces BC-VARETA, a method for estimating brain sources and connectivity from M/EEG data using a frequency domain approach and sparse modeling.

Multiway Canonical Correlation Analysis of Brain Signals

This paper reviews multiway canonical correlation analysis (MCCA) for fusing brain signals across subjects to extract common components, with examples and discussion of its limitations.

Reliable and automatic epilepsy classification with affordable, consumer-grade electroencephalography in rural sub-Saharan Africa

This study uses low-cost consumer-grade EEG and machine learning to classify epilepsy in rural settings, achieving moderate cross-country performance.

The Decision Decoding ToolBOX (DDTBOX) - A multivariate pattern analysis toolbox for event-related potentials

DDTBOX is an open-source toolbox for multivariate pattern analysis of EEG data, enabling classification and regression of experimental variables through spatiotemporal decoding.

Deep Learning Based BCI Control of a Robotic Service Assistant Using Intelligent Goal Formulation

This paper presents a BCI system that enables paralyzed users to control a robotic assistant using thoughts, demonstrating adaptability and task completion in real-world scenarios.

Comparing EEG/MEG neuroimaging methods based on localization error, false positive activity, and false positive connectivity

The paper compares EEG/MEG methods for localization error, false positive activity, and false positive connectivity, finding that sLORETA and eLORETA have exact localization while others perform worse.

Conservative method for attenuation of Vertical Electrooculogram based on local suppression of artifact templates from ongoing EEG

The paper introduces a method to reduce VEOG artifacts in EEG signals by using blink templates and a threshold correlation, effectively recovering ERP and EEG signals with minimal distortion.

Robust detrending, rereferencing, outlier detection, and inpainting for multichannel data

The paper introduces techniques for preprocessing multichannel data to remove artifacts like drift, glitches, and ringing, improving data quality without discarding corrupted trials.

Heart-brain interactions in the MR environment: characterization of the ballistocardiogram in EEG signals collected during simultaneous fMRI

This study characterizes the ballistocardiogram (BCG) artifact in EEG signals during simultaneous fMRI, identifying its temporal and spatial dynamics and proposing insights for improved artifact removal.

A Novel, Fast and Efficient Single-Sensor Automatic Sleep-Stage Classification Based on Complementary Cross-Frequency Coupling Estimates

This paper introduces a single-sensor automatic sleep-stage classification method using cross-frequency coupling estimates, achieving high accuracy and outperforming previous studies.

Decoding the auditory brain with canonical component analysis

This paper introduces Canonical Correlation Analysis (CCA) to decode auditory brain responses and improve stimulus-response mapping for BCI applications.

Task-Related EEG Source Localization via Graph Regularized Low-Rank Representation Model

This paper proposes a novel EEG source localization method that uses a graph-regularized low-rank representation model to improve task-related source reconstruction by eliminating spurious signals.

Sequence Transfer Learning for Neural Decoding

This paper demonstrates the use of LSTM networks for neural decoding in BCIs, showing improved accuracy and robustness compared to traditional linear models.

Taking off the training wheels: Measuring auditory P3 during outdoor cycling using an active wet EEG system

This study measures auditory P3 during outdoor cycling using mobile EEG, finding similar ERP topography and morphology but with reduced amplitude and changes in P2 and N1 components.

A cryptography-based approach for movement decoding

This paper introduces a new decoding method called distribution alignment decoding (DAD) that uses movement statistics to map neural activity to motor variables, achieving comparable or better performance than traditional supervised decoders in some cases.

Non-invasive laminar inference with MEG: Comparison of methods and source inversion algorithms

This paper compares MEG source inversion algorithms and fit metrics to determine the laminar origin of neural activity, finding that sparsity-constrained methods and specific fit metrics improve discrimination accuracy.

Tracking and optimizing human performance using deep reinforcement learning in closed-loop behavioral- and neurofeedback: a proof of concept

This paper applies deep reinforcement learning to improve human sustained attention in a neurofeedback setting using simulated and real EEG data.

Very high density EEG elucidates spatiotemporal aspects of early visual processing

This study compares high-density EEG (SND) with standard density (ND) in capturing spatiotemporal aspects of early visual processing, finding SND provides more accurate and detailed neural information.

Online detection of auditory attention with mobile EEG: closing the loop with neurofeedback

This paper investigates online auditory attention detection with neurofeedback, showing that training can improve performance and that stimuli correlate with AAD outcomes.

MatClassRSA: A Matlab toolbox for M/EEG classification and visualization of proximity matrices

MatClassRSA is a Matlab toolbox for M/EEG classification and visualization of proximity matrices, offering tools for classification, RDM computation, and clustering.

Prediction of Epileptic Seizures based on Mean Phase Coherence

This study presents a real-time technique using Mean Phase Coherence to predict epileptic seizures with a prediction horizon of 4-10 minutes, achieving success in 8 out of 10 test cases.

Pitfalls in the Assessment of Brain-Machine Interfaces Using Information Transfer Rate

The paper identifies flaws in the use of information transfer rate (ITR) to evaluate brain-machine interfaces (BMIs) and discusses the limitations of ITR as a metric for BMI performance.

Four-Sphere Head Model For EEG Signals Revisited

The paper derives and corrects analytical formulas for scalp potentials in the four-sphere EEG head model and compares them with FEM simulations.

Complexity-Entropy maps as a tool for the characterization of the clinical electrophysiological evolution of patients under pharmacological treatment with psychotropic drugs

This paper introduces a complexity-entropy map method to analyze EEG data from patients on psychotropic drugs, showing it can detect subtle changes in clinical response to treatment.

Using spatiotemporal source separation to identify prominent features in multichannel data without sinusoidal filters

This paper introduces a spatiotemporal source separation method to improve signal-to-noise ratio and reduce dimensionality in multichannel neural data without using sinusoidal filters.

Innovations orthogonalization: a solution to the major pitfalls of EEG/MEG "leakage correction"

The paper introduces a new method for unmixing EEG/MEG inverse solution signals to correct for false connectivity, showing it produces accurate human connectomes under broad conditions.

Comparative Analysis of Non-linear Behaviour with Power Spectral Intensity Response Between Normal and Epileptic EEG Signals

This paper compares linear and nonlinear EEG analysis methods to detect differences between normal and epileptic signals using DFA and PSI.

Enhancing the Representational Similarity Between Execution and Imagination of Movement Using Network-Based Brain Computer Interfacing

This paper introduces a neuroconnectivity-based paradigm to enhance the similarity between movement imagination and execution in MI-BCI, showing improved session-to-session consistency in four out of five participants.