Automated preprocessing of 64 channel electroenchephalograms recorded by biosemi instruments

Preprocessing is a mandatory step in electroencephalogram (EEG) signal analysis. Overcoming challenges posed by high noise levels and substantial amplitude artifacts, such as blink-induced electrooculogram (EOG) and muscle-related electromyogram (EMG) interference, is imperative. The signal-to-noise ratio significantly influences the reliability and statistical significance of subsequent analyses. Existing referencing approaches employed in multi-card systems, like using a single electrode or averaging across multiple electrodes, fall short in this respect. In this article, we introduce an innovative referencing method tailored to multi-card instruments, enhancing signal fidelity and analysis outcomes. Our proposed signal processing loop not only mitigates blink-related artifacts but also accurately identifies muscle activity. This work contributes to advancing EEG analysis by providing a robust solution for artifact removal and enhancing data integrity. • Removes blink; • Marks muscle activity; • Re-references with design specific enhancements.