Related Experiment Video
Updated: Jul 30, 2025
06:34
A Single-Channel and Non-Invasive Wearable Brain-Computer Interface for Industry and Healthcare
Published on: July 7, 2023
2.4K
A Systematic Review of Using Deep Learning Technology in the Steady-State Visually Evoked Potential-Based
A S Albahri1, Z T Al-Qaysi2, Laith Alzubaidi3,4
1Iraqi Commission for Computers and Informatics (ICCI), Baghdad, Iraq.
Summary
Deep learning significantly enhances steady-state visually evoked potential (SSVEP) brain-computer interfaces (BCIs). This review analyzes deep learning methods, challenges, and proposes trust solutions for SSVEP-BCI applications.
Area of Science:
- Neuroscience
- Computer Science
- Biomedical Engineering
Background:
- Brain-computer interfaces (BCIs) translate brain activity into commands.
- Steady-state visually evoked potentials (SSVEPs) are a common BCI signal.
- Deep learning (DL) offers advanced pattern recognition for BCIs.
Purpose of the Study:
- To systematically review DL techniques in SSVEP-BCI applications.
- To analyze DL methods, challenges, and trust in SSVEP-BCIs.
- To provide recommendations for researchers and developers.
Main Methods:
- Systematic literature review of PubMed, ScienceDirect, and IEEE (2010-2021).
- Filtering 125 papers to 30 relevant articles.
- Classification of DL methods (CNN, RNN, DNN, LSTM, RBM) and analysis of key aspects.
Main Results:
- Convolutional Neural Networks (CNNs) dominate SSVEP-BCI research (70%).
- Key aspects analyzed include feature extraction, classification, and accuracy.
- Identified challenges in trustworthy DL for SSVEP-BCIs.
Conclusions:
- DL significantly impacts SSVEP-BCI performance.
- Addressing trust and development challenges is crucial.
- A fuzzy decision-making approach is proposed for benchmarking SSVEP-BCIs.
