Title:
Development Of Hybrid Deep Learning Models with Attention Mechanism For Eeg-fnirs in Motor Imagery Classification
Author:
Supervisor:
Department:
Applied Science
Year:
2025
URL:
Abstract:
Brain-Computer Interface (BCI) systems are a transformative advancement in assistive technologies, neurorehabilitation, and cognitive monitoring, enabling direct communication between the brain and ex...
View more
Multimedia Content:
01
Methodology
1. Research Framework
Goal: Develop attention-based hybrid deep learning models for Motor Imagery (MI) classification using EEG, fNIRS, and combined EEG-fNIRS signals.
Stages:
1. Data Acquisiti...
View more
02
Citation Analysis
1. Sensors (MDPI)
Scope: Covers sensors and signal processing, including brain-computer interface (BCI) systems, EEG/fNIRS signal analysis, and machine learning applications.
Relevance: Frequently...
View more
03
Impact
1. Academic Impact
Advances in Computational Neuroscience: Introduces attention-based hybrid deep learning architectures (Bi-LSTM, Bi-GRU, Bi-LSTM-FCN, Bi-LSTM-GRU) for multimodal neuroimaging.
Be...
View more
04
Benefits to Society
1. Scientific & Technical Benefits
Improved Accuracy: Attention-based hybrid models significantly outperform traditional ML and unimodal DL approaches, achieving up to 95% accuracy with EEG-fNIRS f...
View more
05
Applications
1. Motor Imagery-Based Neurorehabilitation
Application: Assisting stroke patients in regaining motor function through BCI-controlled rehabilitation devices.
Use Case: Decoding left/right-hand moto...
View more
06
Infographic
Summary of this research
07
Patents / Products
1. Hybrid EEG-fNIRS BCI Classification Method
Patent: A method for classifying motor imagery tasks using attention-based hybrid deep learning models on EEG and fNIRS signals.
Application: Accurate...
View more
08
Experts
1. Assoc. Prof. Dr. Ibrahima Faye
Institution: Universiti Teknologi PETRONAS (UTP), Malaysia
Contribution: Main supervisor for the thesis; expert in computational neuroscience, signal processing, ...
View more