Recognition and reconstruction of late auditory evoked potentials using wavelet analysis

We introduce a recognition and reconstruction method for late auditory evoked potentials (AEPs) using wavelet analysis. The AEPs are part of the electroencephalogram (EEG) in reaction to acoustic stimuli and are used for diagnostic purposes. The decomposition of the EEG measurement data into subbands can yield a parameterization of the AEP using a small number of significant coefficients, which can be determined by surveying distributions of the transform coefficients. This method is more reliable than traditional means, and additionally offers the possibility of reconstructing potential AEPs by performing an inverse transform on significant coefficients.