The history, current state and future possibilities of the non-invasive brain computer interfaces

Medicine in Novel Technology and Devices, Год журнала: 2025, Номер 25, С. 100353 - 100353

Опубликована: Фев. 1, 2025

Язык: Английский

---

CLTNet: A Hybrid Deep Learning Model for Motor Imagery Classification

Brain Sciences, Год журнала: 2025, Номер 15(2), С. 124 - 124

Опубликована: Янв. 27, 2025

Background: Brain–computer interface (BCI) technology opens up new avenues for human–machine interaction and rehabilitation by connecting the brain to machines. Electroencephalography (EEG)-based motor imagery (MI) classification is a key component of BCI technology, which capable translating neural activity in into commands controlling external devices. Despite great potential challenges extracting decoding signals limit its wide application. Methods: To address this challenge, study proposes novel hybrid deep learning model, CLTNet, focuses on solving feature extraction problem improve MI-EEG signals. In preliminary stage, CLTNet uses convolutional network (CNN) extract time series, channel, spatial features EEG obtain important local information. model combines long short-term memory (LSTM) Transformer module capture time-series data global dependencies EEG. The LSTM explains dynamics activity, while Transformer’s self-attention mechanism reveals series. Ultimately, classifies through fully connected layer. Results: achieved an average accuracy 83.02% Kappa value 0.77 IV 2a dataset, 87.11% 0.74 2b both outperformed traditional methods. Conclusions: innovation that it integrates multiple architectures, offers more comprehensive understanding characteristics during imagery, providing perspective establishing benchmark future research area.

Язык: Английский

---

Data Constraints and Performance Optimization for Transformer-Based Models in EEG-Based Brain-Computer Interfaces: A Survey

IEEE Access, Год журнала: 2024, Номер 12, С. 62628 - 62647

Опубликована: Янв. 1, 2024

This work reviews the critical challenge of data scarcity in developing Transformer-based models for Electroencephalography (EEG)-based Brain-Computer Interfaces (BCIs), specifically focusing on Motor Imagery (MI) decoding. While EEG-BCIs hold immense promise applications communication, rehabilitation, and human-computer interaction, limited availability hinders use advanced deep-learning such as Transformers. In particular, this paper comprehensively analyzes three key strategies to address scarcity: augmentation, transfer learning, inherent attention mechanisms Data augmentation techniques artificially expand datasets, enhancing model generalizability by exposing them a wider range signal patterns. Transfer learning utilizes pre-trained from related domains, leveraging their learned knowledge overcome limitations small EEG datasets. By thoroughly reviewing current research methodologies, underscores importance these overcoming scarcity. It critically examines imposed datasets showcases potential solutions being developed challenges. comprehensive survey, intersection technological advancements, aims provide analysis state-of-the-art EEG-BCI development. identifying gaps suggesting future directions, encourages further exploration innovation field. Ultimately, contribute advancement more accessible, efficient, precise systems addressing fundamental

Язык: Английский

---

Automatic channel selection using multi-objective prioritized jellyfish search (MPJS) algorithm for motor imagery classification using modified DB-EEGNET

Neural Computing and Applications, Год журнала: 2025, Номер unknown

Опубликована: Янв. 23, 2025

Язык: Английский

---

EEG classification with limited data: A deep clustering approach

Pattern Recognition, Год журнала: 2024, Номер 157, С. 110934 - 110934

Опубликована: Авг. 30, 2024

Язык: Английский

---

Decoding Brain Signals from Rapid-Event EEG for Visual Analysis Using Deep Learning

Sensors, Год журнала: 2024, Номер 24(21), С. 6965 - 6965

Опубликована: Окт. 30, 2024

The perception and recognition of objects around us empower environmental interaction. Harnessing the brain's signals to achieve this objective has consistently posed difficulties. Researchers are exploring whether poor accuracy in field is a result design temporal stimulation (block versus rapid event) or inherent complexity electroencephalogram (EEG) signals. Decoding perceptive signal responses subjects become increasingly complex due high noise levels nature brain activities. EEG have resolution non-stationary signals, i.e., their mean variance vary overtime. This study aims develop deep learning model for decoding subjects' rapid-event visual stimuli highlights major factors that contribute low classification task.The proposed multi-class, multi-channel integrates feature fusion handle complex, applied largest publicly available dataset consisting 40 object classes, with 1000 images each class. Contemporary state-of-the-art studies area investigating large number classes achieved maximum 17.6%. In contrast, our approach, which Multi-Class, Multi-Channel Feature Fusion (MCCFF), achieves 33.17% classes. These results demonstrate potential advancing offering future applications machine models.

Язык: Английский

---

Motor imagery EEG signal classification based on deformable convolution v3 and adaptive spatial attention mechanism

Biomedical Signal Processing and Control, Год журнала: 2024, Номер 99, С. 106905 - 106905

Опубликована: Сен. 18, 2024

Язык: Английский

---

Boosted Harris Hawks Shuffled Shepherd Optimization Augmented Deep Learning based motor imagery classification for brain computer interface

PLoS ONE, Год журнала: 2024, Номер 19(11), С. e0313261 - e0313261

Опубликована: Ноя. 21, 2024

Motor imagery (MI) classification has been commonly employed in making brain-computer interfaces (BCI) to manage the outside tools as a substitute neural muscular path. Effectual MI BCI improves communication and mobility for people with breakdown or motor damage, delivering bridge between brain’s intentions exterior actions. Employing electroencephalography (EEG) aggressive recordings, machine learning (ML) methods are used interpret patterns of brain action linked image tasks. These models frequently depend upon like support vector (SVM) deep (DL) distinguish among dissimilar classes, such visualizing left right limb This procedure allows individuals, particularly those disabilities, utilize their opinions command devices robotic limbs computer borders. article presents Boosted Harris Hawks Shuffled Shepherd Optimization Augmented Deep Learning (BHHSHO-DL) technique based on Imagery Classification BCI. The BHHSHO-DL mainly exploits hyperparameter-tuned DL approach identification Initially, performs data preprocessing utilizing wavelet packet decomposition (WPD) model. Besides, enhanced densely connected networks (DenseNet) model extracts preprocessed data’s complex hierarchical feature patterns. Meanwhile, BHHSHO technique-based hyperparameter tuning process is accomplished elect optimal parameter values DenseNet Finally, implemented by convolutional autoencoder (CAE) simulation value methodology performed benchmark dataset. performance validation portrayed superior accuracy 98.15% 92.23% over other techniques under BCIC-III BCIC-IV datasets.

Язык: Английский

---

Lower limb movement state pattern recognition based on EEG-EMG signals

Опубликована: Сен. 13, 2024

Язык: Английский

---