APPLEE: Automated Preprocessing Pipeline for Low-Electrode Encephalography
Luisa M. Zapata S., Juliana Moreno R., E. G., Santiago Caro González, David F. Aguillón N., Carlos Tobón, John F. Ochoa G. · 2024
Electroencephalography (EEG) is crucial for studying cortical neuronal activity in medical diagnostics and research. However, artifacts often contaminate low-density EEG recordings, necessitating careful preprocessing to ensure data reliability. This study introduces APPLEE (Automated Preprocessing Pipeline for Low-Electrode Encephalography), a novel preprocessing pipeline tailored to enhance data in low-density EEG recordings. APPLEE incorporates advanced techniques for artifact reduction, evaluated using data from two cohorts: high-density and low-density (8 electrodes). The results indicate that APPLEE allows the processing of reduced samples with as few as 8 high-density electrodes while preserving signal quality. Additionally, smaller standard deviations were observed in spectral estimations compared to the HAPPILEE pipeline (Harvard Automated Pre-processing Pipeline Include Low-Electrode Encephalography). This comparative analysis underscores APPLEE's innovative methodologies in addressing challenges specific to low-density EEG configurations, demonstrating improved performance in attenuating unwanted frequencies.