In recent years, Machine Learning (ML) has revolutionized numerous scientific and engineering disciplines, and its impact on the analysis and interpretation of electroencephalography (EEG) signals has been particularly profound. Traditionally, EEG signal processing relied heavily on conventional methods characterized by handcrafted features and rule-based logic. These approaches often involved extensive domain expertise to manually design filters, extract specific frequency bands (e.g., alpha, beta, theta), and define thresholds or patterns indicative of brain states or neurological events. While effective to a degree, these traditional methods could be rigid, labor-intensive, and sometimes struggled to capture the subtle, complex, and highly non-linear dynamics inherent in brain activity.