A METHOD OF INDEPENDENT COMPONENTS FOR CLEANING ELECTROENCEPHALOGRAMS FROM ARTIFACTS OF VARIOUS NATURES

View at publisher Summarise with VIRAT
Authors: S.V. Kirillova, А.Г. Зотин, K.V. Simonov
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The study is devoted to solving a pressing problem in digital processing of biomedical signals: cleaning multichannel electroencephalograms (EEG) from artifacts using the independent component method. The aim of the work is to develop algorithmic support for the independent component method for cleaning EEG signals from artifacts and identifying hidden sources of bioelectrical activity. As part of the work, a fragment of a multichannel EEG recording containing a pronounced oculographic artifact was modeled. Based on the FastICA algorithm, the signal was decomposed, the contribution of each component was visualized, and filtering was performed by means of inverse reconstruction of the recording with the exclusion of the noise component. The data preprocessing procedures required for implementing the algorithm are discussed. Experiments are conducted to decompose a generated, artifact-contaminated multichannel EEG recording into independent components, as well as to identify artifacts and reconstruct the cleaned signal. An increase in the signal-to-noise ratio (SNR) from –1.25 dB to 5.23 dB was achieved, which indicates high-quality restoration of the useful signal. The obtained results allow us to conclude that the developed algorithmic software can be used for efficient data processing in neural interface systems and medical diagnostics.

Fields of Study & Subjects

EEG and Brain-Computer Interfaces Blind Source Separation Techniques ECG Monitoring and Analysis

Publication details

Journal / source
Informatization and communication
Volume
1
Publication date
1 January 2026
Language
EN