Independent component analysis of multichannel auditory evoked potentials

We used independent component analysis (ICA) to extract a clean response from multichannel recordings of auditory evoked potentials (AEPs) which represent the electrical response of the brain to auditory stimuli. AEPs are often contaminated by biological noise, such as eye blinks and muscle activity, as well as by extraneous interference. ICA is used to extract the activity of neuronal generators from the noisy recordings. Simulation results show that the method can recover a variety of signals when only noise-contaminated mixtures of these signals are observed. Preliminary results with real data obtained from normal subjects demonstrate that ICA can provide a improved AEP estimate when compared to classical ensemble averaging.