EEG Signal Processing & Brain-Computer Interfaces

ISACLAB develops signal-processing and machine-learning pipelines that turn raw EEG into clinically useful information. We work on motor-imagery classification for brain-computer interfaces, feature fusion and channel-selection methods to boost BCI accuracy, and detection algorithms for epileptic seizures, Parkinson's disease, and locked-in syndrome. Techniques include empirical mode decomposition, multi-domain adaptive feature extraction, multi-layer CNNs, and information-fusion frameworks — published in IEEE Sensors Journal, Cognitive Neurodynamics, and Biomedical Signal Processing and Control. This work extends naturally into related physiological signal analysis, including ECG-based cardiac diagnostics.