Current Status, Challenges, and Possible Solutions of EEG-Based Brain-Computer Interface: A Comprehensive Review

Frontiers in Neurorobotics, Journal Year: 2020, Volume and Issue: 14

Published: June 3, 2020

Brain-Computer Interface (BCI), in essence, aims at controlling different assistive devices through the utilization of brain waves. It is worth noting that application BCI not limited to medical applications, and hence, research this field has gained due attention. Moreover, significant number related publications over past two decades further indicates consistent improvements breakthroughs have been made particular field. Nonetheless, it also mentioning with these improvements, new challenges are constantly discovered. This article provides a comprehensive review state-of-the-art complete system. First, brief overview electroencephalogram (EEG)-based systems given. Secondly, considerable popular applications reviewed terms electrophysiological control signals, feature extraction, classification algorithms, performance evaluation metrics. Finally, recent discussed, possible solutions mitigate issues recommended.

Language: Английский

---

Deep learning techniques for classification of electroencephalogram (EEG) motor imagery (MI) signals: a review

Neural Computing and Applications, Journal Year: 2021, Volume and Issue: 35(20), P. 14681 - 14722

Published: Aug. 25, 2021

Language: Английский

---

Deep learning for motor imagery EEG-based classification: A review

Biomedical Signal Processing and Control, Journal Year: 2020, Volume and Issue: 63, P. 102172 - 102172

Published: Oct. 8, 2020

Language: Английский

---

Investigating the Use of Pretrained Convolutional Neural Network on Cross-Subject and Cross-Dataset EEG Emotion Recognition

Sensors, Journal Year: 2020, Volume and Issue: 20(7), P. 2034 - 2034

Published: April 4, 2020

The electroencephalogram (EEG) has great attraction in emotion recognition studies due to its resistance deceptive actions of humans. This is one the most significant advantages brain signals comparison visual or speech context. A major challenge EEG-based that EEG recordings exhibit varying distributions for different people as well same person at time instances. nonstationary nature limits accuracy it when subject independency priority. aim this study increase subject-independent by exploiting pretrained state-of-the-art Convolutional Neural Network (CNN) architectures. Unlike similar extract spectral band power features from readings, raw data used our after applying windowing, pre-adjustments and normalization. Removing manual feature extraction training system overcomes risk eliminating hidden helps leverage deep neural network's uncovering unknown features. To improve classification further, a median filter eliminate false detections along prediction interval emotions. method yields mean cross-subject 86.56% 78.34% on Shanghai Jiao Tong University Emotion Dataset (SEED) two three classes, respectively. It also 72.81% Database Analysis using Physiological Signals (DEAP) 81.8% Loughborough Multimodal (LUMED) classes. Furthermore, model been trained SEED dataset was tested with DEAP dataset, which 58.1% across all subjects Results show terms accuracy, proposed approach superior to, par with, reference identified literature limited complexity elimination need extraction.

Language: Английский

---

Brain-Computer Interface: Advancement and Challenges

Sensors, Journal Year: 2021, Volume and Issue: 21(17), P. 5746 - 5746

Published: Aug. 26, 2021

Brain-Computer Interface (BCI) is an advanced and multidisciplinary active research domain based on neuroscience, signal processing, biomedical sensors, hardware, etc. Since the last decades, several groundbreaking has been conducted in this domain. Still, no comprehensive review that covers BCI completely yet. Hence, a overview of presented study. This study applications upholds significance Then, each element systems, including techniques, datasets, feature extraction methods, evaluation measurement matrices, existing algorithms, classifiers, are explained concisely. In addition, brief technologies or mostly sensors used BCI, appended. Finally, paper investigates unsolved challenges explains them with possible solutions.

Language: Английский

---

Past, Present, and Future of EEG-Based BCI Applications

Sensors, Journal Year: 2022, Volume and Issue: 22(9), P. 3331 - 3331

Published: April 26, 2022

An electroencephalography (EEG)-based brain-computer interface (BCI) is a system that provides pathway between the brain and external devices by interpreting EEG. EEG-based BCI applications have initially been developed for medical purposes, with aim of facilitating return patients to normal life. In addition initial aim, also gained increasing significance in non-medical domain, improving life healthy people, instance, making it more efficient, collaborative helping develop themselves. The objective this review give systematic overview literature on from period 2009 until 2019. has prepared based three databases PubMed, Web Science Scopus. This was conducted following PRISMA model. review, 202 publications were selected specific eligibility criteria. distribution research domain analyzed further categorized into fields within reviewed domains. equipment used gathering EEG data signal processing methods reviewed. Additionally, current challenges field possibilities future analyzed.

Language: Английский

---

EEG-inception: an accurate and robust end-to-end neural network for EEG-based motor imagery classification

Journal of Neural Engineering, Journal Year: 2021, Volume and Issue: 18(4), P. 046014 - 046014

Published: March 10, 2021

Objective.Classification of electroencephalography (EEG)-based motor imagery (MI) is a crucial non-invasive application in brain-computer interface (BCI) research. This paper proposes novel convolutional neural network (CNN) architecture for accurate and robust EEG-based MI classification that outperforms the state-of-the-art methods.Approach.The proposed CNN model, namely EEG-inception, built on backbone inception-time network, which has showed to be highly efficient time-series classification. Also, an end-to-end classification, as it takes raw EEG signals input does not require complex signal-preprocessing. Furthermore, this data augmentation method enhance accuracy, at least by 3%, reduce overfitting with limited BCI datasets.Main results.The model all methods achieving average accuracy 88.4% 88.6% 2008 Competition IV 2a (four-classes) 2b datasets (binary-classes), respectively. less than 0.025 s test sample suitable real-time processing. Moreover, standard deviation nine different subjects achieves lowest value 5.5 dataset 7.1 dataset, validates robust.Significance.From experiment results, can inferred EEG-inception exhibits strong potential subject-independent classifier tasks.

Language: Английский

---

Concurrent fNIRS and EEG for Brain Function Investigation: A Systematic, Methodology-Focused Review

Sensors, Journal Year: 2022, Volume and Issue: 22(15), P. 5865 - 5865

Published: Aug. 5, 2022

Electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) stand as state-of-the-art techniques for non-invasive neuroimaging. On a unimodal basis, EEG has poor spatial resolution while presenting high temporal resolution. In contrast, fNIRS offers better resolution, though it is constrained by its One important merit shared the that both modalities have favorable portability could be integrated into compatible experimental setup, providing compelling ground development of multimodal fNIRS-EEG integration analysis approach. Despite growing number studies using concurrent designs reported in recent years, methodological reference past remains unclear. To fill this knowledge gap, review critically summarizes status methods currently used studies, an up-to-date overview guideline future projects to conduct studies. A literature search was conducted PubMed Web Science through 31 August 2021. After screening qualification assessment, 92 involving data recordings analyses were included final review. Specifically, three categories analyses, including EEG-informed fNIRS-informed parallel identified explained with detailed description. Finally, we highlighted current challenges potential directions research.

Language: Английский

---

Advanced TSGL-EEGNet for Motor Imagery EEG-Based Brain-Computer Interfaces

IEEE Access, Journal Year: 2021, Volume and Issue: 9, P. 25118 - 25130

Published: Jan. 1, 2021

Deep learning technology is rapidly spreading in recent years and has been extensive attempts the field of Brain-Computer Interface (BCI).Though accuracy Motor Imagery (MI) BCI systems based on deep have greatly improved compared with some traditional algorithms, it still a big problem to clearly interpret models.To address issues, this work first introduces popular model EEGNet compares algorithm Filter-Bank Common Spatial Pattern (FBCSP).After that, considers that 1-D convolution can be explained by special Discrete Wavelet Transform (DWT), depthwise similar (CSP) algorithm.Therefore, improves using Temporary Constrained Sparse Group Lasso (TCSGL) enhance its performance.The proposed TSGL-EEGNet tested Competition IV 2a III IIIa datasets both are 4-classes classification MI tasks.The testing results show achieved 78.96% (0.7194) average (kappa) dataset 2a, which greater than EEGNet, C2CM, MB3DCNN, SS-MEMDBF FBCSP, especially insensitive subjects.The also 85.30% (0.8040) IIIa, MFTFS et al.At last, uses average-validation stacking further effect model.The rates reach 81.34% 88.89%, kappas 0.7511 0.8519 respectively.Additionally, Grad-CAM visualize frequency spatial features learned neural network.

Language: Английский

---

A Multi-Branch Convolutional Neural Network with Squeeze-and-Excitation Attention Blocks for EEG-Based Motor Imagery Signals Classification

Diagnostics, Journal Year: 2022, Volume and Issue: 12(4), P. 995 - 995

Published: April 15, 2022

Electroencephalography-based motor imagery (EEG-MI) classification is a critical component of the brain-computer interface (BCI), which enables people with physical limitations to communicate outside world via assistive technology. Regrettably, EEG decoding challenging because complexity, dynamic nature, and low signal-to-noise ratio signal. Developing an end-to-end architecture capable correctly extracting data's high-level features remains difficulty. This study introduces new model for MI known as Multi-Branch EEGNet squeeze-and-excitation blocks (MBEEGSE). By clearly specifying channel interdependencies, multi-branch CNN attention employed adaptively change channel-wise feature responses. When compared existing state-of-the-art models, suggested achieves good accuracy (82.87%) reduced parameters in BCI-IV2a dataset (96.15%) high gamma dataset.

Language: Английский

---