Abstract:Over recent decades, electroencephalogram (EEG) has become an essential tool in the field of clinical analysis and neurological disease research. However, EEG recordings are notably vulnerable to artifacts during acquisition, especially in clinical settings, which can significantly impede the accurate interpretation of neuronal activity. Blind source separation is currently the most popular method for EEG denoising, but most of the sources it separates often contain both artifacts and brain activity, which may lead to substantial information loss if handled improperly. In this paper, we introduce a dual-threshold denoising method combining spatial filtering with frequency-domain filtering to automatically eliminate electrooculogram (EOG) and electromyogram (EMG) artifacts from multi-channel EEG. The proposed method employs a fusion of second-order blind identification (SOBI) and canonical correlation analysis (CCA) to enhance source separation quality, followed by adaptive threshold to localize the artifact sources, and strict fixed threshold to remove strong artifact sources. Stationary wavelet transform (SWT) is utilized to decompose the weak artifact sources, with subsequent adjustment of wavelet coefficients in respective frequency bands tailored to the distinct characteristics of each artifact. The results of synthetic and real datasets show that our proposed method maximally retains the time-domain and frequency-domain information in the EEG during denoising. Compared with existing techniques, the proposed method achieves better denoising performance, which establishes a reliable foundation for subsequent clinical analyses.

Blind source separation of multichannel electroencephalogram based on wavelet transform and ICA

BLIND SIGNAL SEPARATION OF MULTI-CHANNEL EEG

A Nueral Network Method of Blind Signal Separation Based on Wavelet Transform

Blind Source Separation By Ica For Eeg Multiple Sources Localization

Blind Source Separation Based On Ica

Blind Separation of Biosignals Using Independent Component Analysis

Removing Artifacts and Extracting Patterns in EEG Based on ICA

A Blind Source Separation Algorithm Based on Whitening and Non-linear Decorrelation

[Eliminating Artifacts of EEG Data Based on Independent Component Analysis].

Extracting Epileptic Feature Spikes Using Independent Component Analysis

An iterative algorithm of independent component analysisand the experiment results

Feature Extraction of Brain-Computer Interface Based on Second Order Blind Identification and ICA

Independent component analysis for human epileptic spikes extraction

Engineering Signals' Blind Source Separation in Frequency Domain and Its Application

The Denoising of ECG Signal Based on SCA and Wavelet Packet Transformation

Single-channel Blind-Source Separation Algorithm based on Wavelet Packet Decomposition

Blind images separation based on the subband independent component analysis

EEG emotion recognition based on blind source separation and wavelet

Research on the Methods for Electroencephalogram Feature Extraction Based on Blind Source Separation

Removal of Ocular and Muscular Artifacts from Multi-channel EEG Using Improved Spatial-Frequency Filtering

Algorithm of Blind Source Separation Combined with Medium Filtering