Adaptive transfer learning-based multiscale feature fused deep convolutional neural network for EEG MI multiclassification in brain–computer interface

Engineering Applications of Artificial Intelligence, Journal Year: 2022, Volume and Issue: 116, P. 105347 - 105347

Published: Aug. 30, 2022

Language: Английский

---

A Model Combining Multi Branch Spectral-Temporal CNN, Efficient Channel Attention, and LightGBM for MI-BCI Classification

IEEE Transactions on Neural Systems and Rehabilitation Engineering, Journal Year: 2023, Volume and Issue: 31, P. 1311 - 1320

Published: Jan. 1, 2023

Accurately decoding motor imagery (MI) brain-computer interface (BCI) tasks has remained a challenge for both neuroscience research and clinical diagnosis. Unfortunately, less subject information low signal-to-noise ratio of MI electroencephalography (EEG) signals make it difficult to decode the movement intentions users. In this study, we proposed an end-to-end deep learning model, multi-branch spectral-temporal convolutional neural network with channel attention LightGBM model (MBSTCNN-ECA-LightGBM), MI-EEG tasks. We first constructed multi branch CNN module learn domain features. Subsequently, added efficient mechanism obtain more discriminative Finally, was applied multi-classification The within-subject cross-session training strategy used validate classification results. experimental results showed that achieved average accuracy 86% on two-class MI-BCI data 74% four-class data, which outperformed current state-of-the-art methods. MBSTCNN-ECA-LightGBM can efficiently spectral temporal EEG, improving performance MI-based BCIs.

Language: Английский

---

Multiclass classification of motor imagery tasks based on multi-branch convolutional neural network and temporal convolutional network model

Cerebral Cortex, Journal Year: 2024, Volume and Issue: 34(2)

Published: Jan. 5, 2024

Abstract Motor imagery (MI) is a cognitive process wherein an individual mentally rehearses specific movement without physically executing it. Recently, MI-based brain–computer interface (BCI) has attracted widespread attention. However, accurate decoding of MI and understanding neural mechanisms still face huge challenges. These seriously hinder the clinical application development BCI systems based on MI. Thus, it very necessary to develop new methods decode tasks. In this work, we propose multi-branch convolutional network (MBCNN) with temporal (TCN), end-to-end deep learning framework multi-class We first used MBCNN capture electroencephalography signals information spectral domains through different kernels. Then, introduce TCN extract more discriminative features. The within-subject cross-session strategy validate classification performance dataset Competition IV-2a. results showed that achieved 75.08% average accuracy for 4-class task classification, outperforming several state-of-the-art approaches. proposed MBCNN-TCN-Net successfully captures features decodes tasks effectively, improving MI-BCIs. Our findings could provide significant potential systems.

Language: Английский

---

EEG Signal Complexity Measurements to Enhance BCI-Based Stroke Patients’ Rehabilitation

Sensors, Journal Year: 2023, Volume and Issue: 23(8), P. 3889 - 3889

Published: April 11, 2023

The second leading cause of death and one the most common causes disability in world is stroke. Researchers have found that brain–computer interface (BCI) techniques can result better stroke patient rehabilitation. This study used proposed motor imagery (MI) framework to analyze electroencephalogram (EEG) dataset from eight subjects order enhance MI-based BCI systems for patients. preprocessing portion comprises use conventional filters independent component analysis (ICA) denoising approach. Fractal dimension (FD) Hurst exponent (Hur) were then calculated as complexity features, Tsallis entropy (TsEn) dispersion (DispEn) assessed irregularity parameters. features statistically retrieved each participant using two-way variance (ANOVA) demonstrate individuals’ performances four classes (left hand, right foot, tongue). dimensionality reduction algorithm, Laplacian Eigenmap (LE), was classification performance. Utilizing k-nearest neighbors (KNN), support vector machine (SVM), random forest (RF) classifiers, groups post-stroke patients ultimately determined. findings show LE with RF KNN obtained 74.48% 73.20% accuracy, respectively; therefore, integrated set along ICA technique exactly describe MI framework, which may be explore will help clinicians, doctors, technicians make a good rehabilitation program people who had

Language: Английский

---

A subject-independent portable emotion recognition system using synchrosqueezing wavelet transform maps of EEG signals and ResNet-18

Biomedical Signal Processing and Control, Journal Year: 2023, Volume and Issue: 90, P. 105875 - 105875

Published: Dec. 30, 2023

Language: Английский

---

Recent Trends in EEG-Based Motor Imagery Signal Analysis and Recognition: A Comprehensive Review

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 80518 - 80542

Published: Jan. 1, 2023

The electroencephalogram (EEG) motor imagery (MI) signals are the widespread paradigms in brain-computer interface (BCI). Its significant applications gaming, robotics, and medical fields drew our attention to perform a detailed analysis. However, problem is ill-posed as these highly nonlinear, unpredictable, noisy, hence making it exceedingly hard be analyzed adequately. This paper provides first-of-its-kind comprehensive review of conventional signal processing deep learning techniques for BCI MI comprises extensive works carried out domain recent past, highlighting current challenges problem. A new categorization existing approaches has been presented better clarification. An all-inclusive description corroborated by relevant area. Moreover, architectures various standard algorithms along with their merits demerits also explicated assist readers. tabular representations numerical results readily provided. work presents open research problems future directions.

Language: Английский

---

ETCNet: An EEG-based motor imagery classification model combining efficient channel attention and temporal convolutional network

Brain Research, Journal Year: 2023, Volume and Issue: 1823, P. 148673 - 148673

Published: Nov. 11, 2023

Language: Английский

---

Deep learning in motor imagery EEG signal decoding: A Systematic Review

Neurocomputing, Journal Year: 2024, Volume and Issue: 610, P. 128577 - 128577

Published: Sept. 14, 2024

Language: Английский

---

Application of Transfer Learning for Biomedical Signals: A Comprehensive Review of the Last Decade (2014-2024)

Information Fusion, Journal Year: 2025, Volume and Issue: 118, P. 102982 - 102982

Published: Jan. 30, 2025

Language: Английский

---

A novelty towards neural signatures − Unveiling the inter-subject distance metric for EEG-based motor imagery

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 105, P. 107552 - 107552

Published: Feb. 6, 2025

Language: Английский

---