.png){kind=icon}
.png){kind=icon}
Return
Motion Artifact Removal from EEG Signals Using the Motion-Net Deep Learning Algorithm
.png){kind=icon}
OA
AI
DOI：10.1016/j.bspc.2025.108877.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Development of Motion-Net, a CNN-based deep learning model for subject-specific motion artifact removal from EEG signals. • First study to apply visibility graph (VG) features for EEG motion artifact removal, enhancing model accuracy on smaller datasets. • Evaluation of Motion-Net on a dataset with real-world motion artifacts, demonstrating an artifact reduction percentage ( η) of 86% ±4.13 and an SNR improvement of 20 ±4.47 dB. • Motion-Net’s performance validated across three experimental setups, showing that separate encoding of VG features improves artifact removal consistency and signal integrity. • Potential application of Motion-Net in mobile EEG setups for improved data quality in naturalistic settings.
Keywords:
Mobile EEG
Motion artifact removal
U-net
EEG signal analysis
Deep learning
Visibility graph
Feature selection
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4.9
Papers:
9.8K
Citations:
2.4W
Organization
No organization information available