Automated analysis of EEG: opportunities and pitfalls

Automated analysis is the transformation and representation of raw EEG data in an alternate form to allow clinicians to better or more quickly understand and interpret the data for diagnostic purposes. For the purposes of development, automated analysis encompasses everything from simple transforms such as the Fast Fourier Transform for isolating spectral components of EEG to full binary detectors. These analyses can help clinicians detect and diagnose clinical conditions such as seizures, traumatic brain injuries, and many other types of disease. In this article, opportunities and hazards will be examined as seen from the prospective of the automated analysis developer.