IC-U-Net: A U-Net-based Denoising Autoencoder Using Mixtures of Independent Components for Automatic EEG Artifact Removal

Augmenting brain-computer interfaces with ART: An artifact removal transformer for reconstructing multichannel EEG signals

Artifact removal in electroencephalography (EEG) is a longstanding challenge that significantly impacts neuroscientific analysis and brain-computer interface (BCI) performance. Tackling this problem demands advanced algorithms, extensive noisy-clean training data, and thorough evaluation strategies. This study presents the Artifact Removal Transformer (ART), an innovative EEG denoising model employing transformer architecture to adeptly capture the transient millisecond-scale dynamics characteristic of EEG signals. Our approach offers a holistic, end-to-end denoising solution that simultaneously addresses multiple artifact types in multichannel EEG data. We enhanced the generation of noisy-clean EEG data pairs using an independent component analysis, thus fortifying the training scenarios critical for effective supervised learning. We performed comprehensive validations using a wide range of open datasets from various BCI applications, employing metrics like mean squared error and signal-to-noise ratio, as well as sophisticated techniques such as source localization and EEG component classification. Our evaluations confirm that ART surpasses other deep-learning-based artifact removal methods, setting a new benchmark in EEG signal processing. This advancement not only boosts the accuracy and reliability of artifact removal but also promises to catalyze further innovations in the field, facilitating the study of brain dynamics in naturalistic environments.

Brain computer interfaces for cognitive enhancement in older people - challenges and applications: a systematic review

BACKGROUND

Brain-computer interface (BCI) offers promising solutions to cognitive enhancement in older people. Despite the clear progress received, there is limited evidence of BCI implementation for rehabilitation. This systematic review addresses BCI applications and challenges in the standard practice of EEG-based neurofeedback (NF) training in healthy older people or older people with mild cognitive impairment (MCI).

METHODS

Articles were searched via MEDLINE, PubMed, SCOPUS, SpringerLink, and Web of Science. 16 studies between 1st January 2010 to 1st November 2024 are included after screening using PRISMA. The risk of bias, system design, and neurofeedback protocols are reviewed.

RESULTS

The successful BCI applications in NF trials in older people were biased by the randomisation process and outcome measurement. Although the studies demonstrate promising results in effectiveness of research-grade BCI for cognitive enhancement in older people, it is premature to make definitive claims about widespread BCI usability and applicability.

SIGNIFICANCE

This review highlights the common issues in the field of EEG-based BCI for older people. Future BCI research could focus on trial design and BCI performance gaps between the old and the young to develop a robust BCI system that compensates for age-related declines in cognitive and motor functions.

Task-oriented EEG denoising generative adversarial network for enhancing SSVEP-BCI performance

Objective.The quality of electroencephalogram (EEG) signals directly impacts the performance of brain-computer interface (BCI) tasks. Many methods have been proposed to eliminate noise from EEG signals, but most of these methods focus solely on signal denoising itself, disregarding the impact on subsequent tasks, which deviates from the original intention of EEG denoising. The main objective of this study is to optimize EEG denoising models with a purpose of improving the performance of BCI tasks.Approach.To this end, we proposed an innovative task-oriented EEG denoising generative adversarial network (TOED-GAN) method. This network utilizes the generator of GAN to decompose and reconstruct clean signals from the raw EEG signals, and the discriminator to learn to distinguish the generated signals from the true clean signals, resulting in a remarkable increase of the signal-to-noise ratio by simultaneously enhancing task-related components and removing task-irrelevant noise from the original contaminated signals.Main results.We evaluated the performance of the model on a public dataset and a self-collected dataset respectively, with canonical correlation analysis classification tasks of the steady-state visual evoked potential (SSVEP) based BCI. Experimental results demonstrate that TOED-GAN exhibits excellent performance in removing EEG noise and improving performance for SSVEP-BCI, with accuracy improvement rates reaching 18.47% and 21.33% in contrast to the baseline methods of convolutional neural networks, respectively.Significance.This work proves that the proposed TOED-GAN, as an EEG denoising method tailored for SSVEP tasks, contributes to enhancing the performance of BCIs in practical application scenarios.

CA-ViT: Contour-Guided and Augmented Vision Transformers to Enhance Glaucoma Classification Using Fundus Images

Glaucoma, a predominant cause of visual impairment on a global scale, poses notable challenges in diagnosis owing to its initially asymptomatic presentation. Early identification is vital to prevent irreversible vision impairment. Cutting-edge deep learning techniques, such as vision transformers (ViTs), have been employed to tackle the challenge of early glaucoma detection. Nevertheless, limited approaches have been suggested to improve glaucoma classification due to issues like inadequate training data, variations in feature distribution, and the overall quality of samples. Furthermore, fundus images display significant similarities and slight discrepancies in lesion sizes, complicating glaucoma classification when utilizing ViTs. To address these obstacles, we introduce the contour-guided and augmented vision transformer (CA-ViT) for enhanced glaucoma classification using fundus images. We employ a Conditional Variational Generative Adversarial Network (CVGAN) to enhance and diversify the training dataset by incorporating conditional sample generation and reconstruction. Subsequently, a contour-guided approach is integrated to offer crucial insights into the disease, particularly concerning the optic disc and optic cup regions. Both the original images and extracted contours are given to the ViT backbone; then, feature alignment is performed with a weighted cross-entropy loss. Finally, in the inference phase, the ViT backbone, trained on the original fundus images and augmented data, is used for multi-class glaucoma categorization. By utilizing the Standardized Multi-Channel Dataset for Glaucoma (SMDG), which encompasses various datasets (e.g., EYEPACS, DRISHTI-GS, RIM-ONE, REFUGE), we conducted thorough testing. The results indicate that the proposed CA-ViT model significantly outperforms current methods, achieving a precision of 93.0%, a recall of 93.08%, an F1 score of 92.9%, and an accuracy of 93.0%. Therefore, the integration of augmentation with the CVGAN and contour guidance can effectively enhance glaucoma classification tasks.

Challenging Deep Learning Methods for EEG Signal Denoising under Data Corruption

Capturing informative electroencephalogram (EEG) signals is a challenging task due to the presence of noise (e.g., due to human movement). In extreme cases, data recordings from specific electrodes (channels) can become corrupted and entirely devoid of information. Motivated by recent work on deep-learning-based approaches for EEG signal denoising, we present the first benchmark study on the performance of EEG signal denoising methods in the presence of corrupted channels. We design our study considering a wide variety of datasets, models, and evaluation tasks. Our results highlight the need for assessing the performance of EEG deep-learning models across a broad suite of datasets, as provided by our benchmark.

ATN-Res2Unet: an advanced deep learning network for the elimination of saturation artifacts in endoscopy optical coherence tomography

Endoscopic optical coherence tomography (OCT) possesses the capability to non-invasively image internal lumens; however, it is susceptible to saturation artifacts arising from robust reflective structures. In this study, we introduce an innovative deep learning network, ATN-Res2Unet, designed to mitigate saturation artifacts in endoscopic OCT images. This is achieved through the integration of multi-scale perception, multi-attention mechanisms, and frequency domain filters. To address the challenge of obtaining ground truth in endoscopic OCT, we propose a method for constructing training data pairs. Experimental in vivo data substantiates the effectiveness of ATN-Res2Unet in reducing diverse artifacts while preserving structural information. Comparative analysis with prior studies reveals a notable enhancement, with average quantitative indicators increasing by 45.4-83.8%. Significantly, this study marks the inaugural exploration of leveraging deep learning to eradicate artifacts from endoscopic OCT images, presenting considerable potential for clinical applications.

DTP-Net: Learning to Reconstruct EEG Signals in Time-Frequency Domain by Multi-Scale Feature Reuse

Electroencephalography (EEG) signals are prone to contamination by noise, such as ocular and muscle artifacts. Minimizing these artifacts is crucial for EEG-based downstream applications like disease diagnosis and brain-computer interface (BCI). This paper presents a new EEG denoising model, DTP-Net. It is a fully convolutional neural network comprising Densely-connected Temporal Pyramids (DTPs) placed between two learnable time-frequency transformations. In the time-frequency domain, DTPs facilitate efficient propagation of multi-scale features extracted from EEG signals of any length, leading to effective noise reduction. Comprehensive experiments on two public semi-simulated datasets demonstrate that the proposed DTP-Net consistently outperforms existing state-of-the-art methods on metrics including relative root mean square error (RRMSE) and signal-to-noise ratio improvement ( ∆SNR). Moreover, the proposed DTP-Net is applied to a BCI classification task, yielding an improvement of up to 5.55% in accuracy. This confirms the potential of DTP-Net for applications in the fields of EEG-based neuroscience and neuro-engineering. An in-depth analysis further illustrates the representation learning behavior of each module in DTP-Net, demonstrating its robustness and reliability.

Motion Artifact Reduction Using U-Net Model with Three-Dimensional Simulation-Based Datasets for Brain Magnetic Resonance Images

This study aimed to remove motion artifacts from brain magnetic resonance (MR) images using a U-Net model. In addition, a simulation method was proposed to increase the size of the dataset required to train the U-Net model while avoiding the overfitting problem. The volume data were rotated and translated with random intensity and frequency, in three dimensions, and were iterated as the number of slices in the volume data. Then, for every slice, a portion of the motion-free k-space data was replaced with motion k-space data, respectively. In addition, based on the transposed k-space data, we acquired MR images with motion artifacts and residual maps and constructed datasets. For a quantitative evaluation, the root mean square error (RMSE), peak signal-to-noise ratio (PSNR), coefficient of correlation (CC), and universal image quality index (UQI) were measured. The U-Net models for motion artifact reduction with the residual map-based dataset showed the best performance across all evaluation factors. In particular, the RMSE, PSNR, CC, and UQI improved by approximately 5.35×, 1.51×, 1.12×, and 1.01×, respectively, and the U-Net model with the residual map-based dataset was compared with the direct images. In conclusion, our simulation-based dataset demonstrates that U-Net models can be effectively trained for motion artifact reduction.

Parkinson's disease detection from EEG signal employing autoencoder and RBFNN-based hybrid deep learning framework utilizing power spectral density

OBJECTIVE

Early detection of Parkinson's disease (PD) is essential for halting its progression, yet challenges remain in leveraging deep learning for accurate identification. This study aims to overcome these obstacles by introducing a hybrid deep learning approach that enhances PD detection through a combination of autoencoder (AE) and radial basis function neural network (RBFNN).

METHODS

The proposed method analyzes the power spectral density (PSD) of preprocessed electroencephalography (EEG) signals, with artifacts removed, to assess energy distribution across EEG sub-bands. AEs are employed to extract features from reconstructed signals, which are subsequently classified by an RBFNN. The approach is validated on UC SanDiego's EEG dataset, consisting of 31 subjects and 93 minutes of recordings.

RESULTS

The hybrid model demonstrates promising performance, achieving a classification accuracy of 99%. The improved accuracy is attributed to advanced feature selection techniques, robust data preprocessing, and the integration of AEs with RBFNN, setting a new benchmark in PD detection frameworks.

CONCLUSION

This study highlights the efficacy of the hybrid deep learning framework in detecting PD, particularly emphasizing the importance of using multiple EEG channels and advanced preprocessing techniques. The results underscore the potential of this approach for practical clinical applications, offering a reliable solution for early and accurate PD detection.

Emotion recognition in EEG signals using deep learning methods: A review

Emotions are a critical aspect of daily life and serve a crucial role in human decision-making, planning, reasoning, and other mental states. As a result, they are considered a significant factor in human interactions. Human emotions can be identified through various sources, such as facial expressions, speech, behavior (gesture/position), or physiological signals. The use of physiological signals can enhance the objectivity and reliability of emotion detection. Compared with peripheral physiological signals, electroencephalogram (EEG) recordings are directly generated by the central nervous system and are closely related to human emotions. EEG signals have the great spatial resolution that facilitates the evaluation of brain functions, making them a popular modality in emotion recognition studies. Emotion recognition using EEG signals presents several challenges, including signal variability due to electrode positioning, individual differences in signal morphology, and lack of a universal standard for EEG signal processing. Moreover, identifying the appropriate features for emotion recognition from EEG data requires further research. Finally, there is a need to develop more robust artificial intelligence (AI) including conventional machine learning (ML) and deep learning (DL) methods to handle the complex and diverse EEG signals associated with emotional states. This paper examines the application of DL techniques in emotion recognition from EEG signals and provides a detailed discussion of relevant articles. The paper explores the significant challenges in emotion recognition using EEG signals, highlights the potential of DL techniques in addressing these challenges, and suggests the scope for future research in emotion recognition using DL techniques. The paper concludes with a summary of its findings.

An Approach for EEG Denoising Based on Wasserstein Generative Adversarial Network

Electroencephalogram (EEG) recordings often contain artifacts that would lower signal quality. Many efforts have been made to eliminate or at least minimize the artifacts, and most of them rely on visual inspection and manual operations, which is time/labor-consuming, subjective, and incompatible to filter massive EEG data in real-time. In this paper, we proposed a deep learning framework named Artifact Removal Wasserstein Generative Adversarial Network (AR-WGAN), where the well-trained model can decompose input EEG, detect and delete artifacts, and then reconstruct denoised signals within a short time. The proposed approach was systematically compared with commonly used denoising methods including Denoised AutoEncoder, Wiener Filter, and Empirical Mode Decomposition, with both public and self-collected datasets. The experimental results proved the promising performance of AR-WGAN on automatic artifact removal for massive data across subjects, with correlation coefficient up to 0.726±0.033, and temporal and spatial relative root-mean-square error as low as 0.176±0.046 and 0.761±0.046, respectively. This work may demonstrate the proposed AR-WGAN as a high-performance end-to-end method for EEG denoising, with many on-line applications in clinical EEG monitoring and brain-computer interfaces.

Enhancing EEG Artifact Removal Efficiency by Introducing Dense Skip Connections to IC-U-Net

Electroencephalographic (EEG) data is considered contaminated with various types of artifacts. Deep learning has been successfully applied to developing EEG artifact removal techniques to increase the signal-to-noise ratio (SNR) and enhance brain-computer interface performance. Recently, our research team has proposed an end-to-end UNet-based EEG artifact removal technique, IC-U-Net, which can reconstruct signals against various artifacts. However, this model suffers from being prone to overfitting with a limited training dataset size and demanding a high computational cost. To address these issues, this study attempted to leverage the architecture of UNet++ to improve the practicability of IC-U-Net by introducing dense skip connections in the encoder-decoder architecture. Results showed that this proposed model obtained superior SNR to the original model with half the number of parameters. Also, this proposed model achieved comparable convergency using a quarter of the training data size.