PROPOSEDPOD 04: Healthcare & Bio-Signals
Artifact Removal and Spectral Power Tracking in Mobile EEG Feeds
Objective & Abstract
Investigating wavelet decomposition methods for removing ocular and muscle artifacts from low-channel headband recordings without losing spectral band signatures.
Investigated Methodologies
- Discrete Wavelet Transform
- Bandpower spectral density estimation
- Blind source separation
Datasets & Benchmarking Corpora
OpenBMI EEG Benchmark
All training and evaluation datasets undergo ethics audit and licensing compliance per AIAC Research Council standards.