Elevating Neuro-Linguistic Decoding: Deepening Neural-Device Interaction with RNN-GRU for Non-Invasive Language Decoding

International Journal of Advanced Computer Science and Applications, Год журнала: 2024, Номер 15(2)

Опубликована: Янв. 1, 2024

Exploring innovative pathways for non-invasive neural communication with language interfaces, this research delves into the interdisciplinary realm of neurolinguistic learning, merging neuroscience and machine learning. It scrutinizes intricacies decoding patterns associated comprehension. Leveraging advanced network architectures, specifically Deep Recurrent Neural Networks (RNN) Gated Units (GRU), study aims to amplify landscape neuro-device interaction. The focus Neurolinguistic Learning lies in extracting language-related brain signals without resorting invasive procedures. Employing cutting-edge methods deep learning techniques, elevate capabilities devices such as brain-machine interfaces neuroprosthetics. A distinctive approach involves crafting a sophisticated RNN-GRU model designed capture intricate linked processing. This architectural innovation, implemented Python software environment, harnesses strengths RNNs GRUs enhance decoding. study's outcomes hold promise advancing systems, contributing expanding knowledge base remarkable accuracy proposed model, boasting 90% rate, signifies its potential application critical real-world scenarios. includes assistive technologies where precise cerebral is paramount. underscores efficacy methodologies pushing boundaries neurotechnology. Notably, outperforms established surpassing alternatives like CSP-SVM EEGNet by an impressive 30.4% accuracy. model's proficiency deciphering topic words ability extract from inputs.

Язык: Английский

---

Mental Stress Classification Based on Selected Electroencephalography Channels Using Correlation Coefficient of Hjorth Parameters

Brain Sciences, Год журнала: 2023, Номер 13(9), С. 1340 - 1340

Опубликована: Сен. 18, 2023

Electroencephalography (EEG) signals offer invaluable insights into diverse activities of the human brain, including intricate physiological and psychological responses associated with mental stress. A major challenge, however, is accurately identifying stress while mitigating limitations a large number EEG channels. Such encompass computational complexity, potential overfitting, prolonged setup time for electrode placement, all which can hinder practical applications. To address these challenges, this study presents novel CCHP method, aimed at ranking commonly optimal channels based on their sensitivity to state. This method's uniqueness lies in its ability not only find common channels, but also prioritize them according responsiveness stress, ensuring consistency across subjects making it potentially transformative real-world From our rigorous examinations, eight emerged as universally detecting variances participants. Leveraging features from time, frequency, time-frequency domains employing machine learning algorithms, notably RLDA, SVM, KNN, approach achieved remarkable accuracy 81.56% SVM algorithm outperforming existing methodologies. The implications research are profound, offering stepping stone toward development real-time detection devices, consequently, enabling clinicians make more informed therapeutic decisions comprehensive brain activity monitoring.

Язык: Английский

---

Enhancing Classification Accuracy with Integrated Contextual Gate Network: Deep Learning Approach for Functional Near-Infrared Spectroscopy Brain–Computer Interface Application

Sensors, Год журнала: 2024, Номер 24(10), С. 3040 - 3040

Опубликована: Май 10, 2024

Brain–computer interface (BCI) systems include signal acquisition, preprocessing, feature extraction, classification, and an application phase. In fNIRS-BCI systems, deep learning (DL) algorithms play a crucial role in enhancing accuracy. Unlike traditional machine (ML) classifiers, DL eliminate the need for manual extraction. neural networks automatically extract hidden patterns/features within dataset to classify data. this study, hand-gripping (closing opening) two-class motor activity from twenty healthy participants is acquired, integrated contextual gate network (ICGN) algorithm (proposed) applied that enhance classification The proposed extracts features filtered data generates patterns based on information previous cells network. Accordingly, performed similar generated dataset. accuracy of compared with long short-term memory (LSTM) bidirectional (Bi-LSTM). ICGN yielded 91.23 ± 1.60%, which significantly (p < 0.025) higher than 84.89 3.91 88.82 1.96 achieved by LSTM Bi-LSTM, respectively. An open access, three-class (right- left-hand finger tapping dominant foot tapping) 30 subjects used validate algorithm. results show can be efficiently two- problems fNIRS-based BCI applications.

Язык: Английский

---

Differences in Electroencephalography Power Levels between Poor and Good Performance in Attentional Tasks

Brain Sciences, Год журнала: 2024, Номер 14(6), С. 527 - 527

Опубликована: Май 22, 2024

Decreased attentional function causes problems in daily life. However, a quick and easy evaluation method of has not yet been developed. Therefore, we are searching for to evaluate easily quickly. This study aimed collect basic data on the features electroencephalography (EEG) during attention tasks develop new evaluating using EEG. Twenty healthy young adults participated; examined cerebral activity Clinical Assessment Attention portable EEG devices. The Mann–Whitney U test was performed assess differences power levels between low- high-attention groups. findings revealed that group showed significantly higher δ wave L-temporal bilateral parietal lobes, as well β γ waves R-occipital lobe, than did low-attention digit-forward, whereas θ R-frontal α frontal lobes digit-backward. Notably, lower θ, α, bands right hemisphere found may be key elements detect deficit.

Язык: Английский

---

An Introductory Guide on Creating a Pandas-based EEG Analysis and Action Prediction Tool for BCI Systems

2022 IEEE 11th International Conference on Communication Systems and Network Technologies (CSNT), Год журнала: 2024, Номер 13, С. 1372 - 1378

Опубликована: Апрель 6, 2024

Brain computer interfaces (BCIs) are rapidly gaining a lot of momentum within the biomedical engineer's sphere. The BCI is link between brain's electrical activity and device that monitors actions functions based on its input. In this paper, we have created prediction algorithm for systems takes in EEG data (i.e., classified actions) using machine learning (ML) techniques. Furthermore, obtained subsequently examined under specific conditions. This necessary as would otherwise lack significance computation. due to fact mostly consists highly disordered brain wave activity. analysis phase study, many Python libraries could be used ranging from MNE library which an essential tool scikit branches ML. project has special emphasis use Pandas project's been workers interns Turkish government agency called scientific technological research council Türkiye (TÜBİTAK). While was being recorded, recording software assigns condition inputs attach them epoched time data.

Язык: Английский

---