EEG-Based Brain-Computer Interface for Recognition of Human Brain Activity During Yoga and Meditation
Brain-Computer Interfaces for Neurorehabilitation December 12, 2025 Bandari Ramaraju, Ravichander Janapati, Sreedhar Kollem
A proposed EEG-based Brain Computer Interface (BCI) framework aims to recognize brain activity during yoga and meditation by analyzing neural oscillations. The methodology involves acquiring EEG signals from multiple scalp regions during various postures, removing artifacts with Independent Component Analysis, and extracting time-frequency features like band power in delta, theta, alpha, beta, and gamma bands. Principal Component Analysis reduces feature dimensionality, and machine learning algorithms such as Support Vector Machines, Random Forests, and EEGNet classify mental states. Performance is evaluated using accuracy, F1-score, and confusion matrices. The framework could enable real-time neurofeedback for mindfulness and focus, and be integrated into clinical neurorehabilitation and cognitive training systems.