EEG and Brain-Computer Interfaces

Correcting MEG artifacts caused by overt speech

This paper develops an artifact rejection method for MEG data during speech production, showing it effectively removes speech-related distortions.

Neural decoding of gait phase information during motor imagery and improvement of the decoding accuracy by concurrent action observation

This study decodes gait phase information during motor imagery using EEG and shows that combining motor imagery with action observation improves decoding accuracy.

Benefits of Deep Learning Classification of Continuous Noninvasive Brain Computer Interface Control

This study shows deep learning improves BCI classification accuracy and information transfer rate using full scalp EEG data, benefiting users with lower initial performance.

A Neuroscience-Inspired Spiking Neural Network for Auditory Spatial Attention Detection Using Single-Trial EEG

This paper introduces a spiking neural network for detecting auditory spatial attention from single-trial EEG data, achieving 90% accuracy with only 10% of the data.

Automated detection of artefacts in neonatal EEG with residual neural networks

This paper develops a deep learning algorithm to detect and classify artefacts in neonatal EEG signals, achieving high accuracy in identifying artefact types and clean EEG periods.

Generative adversarial neural networks maintain decoder accuracy during signal disruption in simulated long-term recordings from brain computer interfaces

This paper presents a GAN-based method to recover degraded signals in simulated long-term BCIs, maintaining decoder accuracy despite signal disruption.

High-performance brain-to-text communication via imagined handwriting

This paper demonstrates a brain-computer interface that decodes imagined handwriting movements into text at high speed and accuracy, closing the gap between BCI and able-bodied typing rates.

A Novel Tool for the Removal of Muscle Artefacts from EEG: Improving Data Quality in the Gamma Frequency Range

This paper introduces a novel method to remove muscle artefacts from EEG data, improving gamma band signal quality by preserving task-relevant cortical activity.

Investigating the Effects of Macaque Primary Motor Cortex Multi-Unit Activity Binning Period on Behavioural Decoding Performance

This paper examines how varying the binning period of multi-unit activity affects BCI decoding performance in macaques, finding a small but significant decrease in performance with longer binning periods.

EEG-based detection of the locus of auditory attention with convolutional neural networks

This study uses convolutional neural networks to detect auditory attention locus (left/right) from EEG signals in multi-speaker scenarios, achieving 81% median accuracy within 1-2 seconds.

Auditory Stimulus-response Modeling with a Match-Mismatch Task

The paper introduces a match-mismatch task to evaluate auditory stimulus-response models, showing state-of-the-art performance and providing a reference implementation.

Effects of Peripheral Haptic Feedback on Intracortical Brain-Computer Interface Control and Associated Sensory Responses in Motor Cortex

This study investigates how haptic feedback affects motor cortical control and iBCI performance in a tetraplegic participant.

Decoding visual colour from scalp electroencephalography measurements

This study decodes visual colour from scalp EEG using Linear Discriminant Analysis, showing it reflects sensory qualities and is comparable to orientation decoding.

Neural silences can be localized rapidly using noninvasive scalp EEG

SilenceMap detects and localizes neural silences using scalp EEG, outperforming existing methods in accuracy and speed.

Fuzzy-Discernibility Matrix-based Efficient Feature Selection Techniques for Improved Motor-Imagery EEG Signal Classification

The paper proposes a mix-bagging classifier with sliding window feature extraction and fuzzy discernibility matrix for motor-imagery EEG classification, achieving 92.14% accuracy on BCI Competition II Dataset III.

Temporal Window based Feature Extraction Technique for Motor-Imagery EEG Signal Classification

The paper compares temporal window-based and filter-bank approaches for motor-imagery EEG classification, finding that overlapping segments achieve the highest accuracy of 91.43% on BCI Competition II Dataset III.

Mobile ear-EEG to study auditory attention in everyday life

This study uses mobile ear-EEG to examine auditory attention in real-world settings, finding increased P3 amplitude and better performance in a controlled rapid oddball task compared to a sporadic one.

Analysis and Visualization of Sleep Stages based on Deep Neural Networks

This study develops a deep learning method to analyze and visualize sleep stages using EEG data, showing it can estimate sleep cycle lengths and detect disturbances.

Using time causal quantifiers to characterize Sleep stages

The paper uses five information theory-based measures to quantify and classify sleep stages from EEG signals, showing that combined measures improve differentiation of sleep states.

Virtual EEG-electrodes: Convolutional neural networks as a method for upsampling or restoring channels

This paper uses convolutional neural networks to upsample or restore EEG channels, showing better performance than interpolation methods.

Towards Semantic fMRI Neurofeedback: Navigating among Mental States using Real-time Representational Similarity Analysis

This paper introduces a real-time fMRI neurofeedback method using semantic representations to allow participants to navigate mental states by controlling neural activity towards target brain states.

Continental generalization of an AI system for clinical seizure recognition

This paper demonstrates an AI system for seizure detection that generalizes across continents, achieving human-level performance with significantly reduced annotation time.

EEG in Motion: Using an Oddball Task to Explore Motor Interference in Active Skateboarding

This study uses an auditory oddball task during skateboarding to examine motor interference on cognitive resources, finding no modulation of P3 and tonic alpha amplitude with increased riding difficulty.

Effect of number and placement of EEG electrodes onmeasurement of neural tracking of speech

The paper evaluates how electrode number and placement affect neural speech tracking in EEG, finding optimal configurations for both subject-specific and subject-independent setups.

Comparison between traditional FFT and marginal spectra using the Hilbert-Huang transform method for the broadband spectral analysis of the EEG in healthy humans.

This study compares traditional FFT and Hilbert-Huang transform methods for EEG spectral analysis in healthy humans, finding similar performance with some band-specific differences.

High-pass filtering artifacts in multivariate classification of neural time series data

The paper shows that high-pass filtering and standard robust detrending introduce artifacts in multivariate pattern classification of neural time series, and proposes trial-masked robust detrending as an alternative that avoids these issues.

Imagined speech can be decoded from low- and cross-frequency features in perceptual space

The paper decodes imagined speech using low- and cross-frequency neural features from electrocorticography data, showing their effectiveness in perceptual speech spaces.

EPOC outside the shield: comparing the performance of a consumer-grade EEG device in shielded and unshielded environments.

The paper compares the performance of the EPOC EEG device in shielded and unshielded environments, finding no significant difference in N170 ERP measurements.

Robust and accurate decoding of hand kinematics from entire spiking activity using deep learning

This paper develops a deep learning method for decoding hand kinematics from neural signals, achieving high accuracy and robustness over long-term recordings in non-human primates.

Distributed source modeling of intracranial stereoelectroencephalographic measurements

This paper develops a distributed sEEG source-modeling approach for analyzing intracranial recordings, showing it effectively captures variance and provides consistent results for event-related potentials and high broadband gamma activity.