Examining EEG signals with parametric and non-parametric analyses methods in migraine patients during pregnancy

EEG tools are used for diagnosing and interpreting brain diseases at neurology clinics. In this study, EEG data collected from healthy, migraine, and pregnant women were analyzed using Modified Covariance (a parametric spectral analysis method) and Welch (a non-parametric method) methods, in an attempt to re-evaluate the value EEG in migraine diagnosis. We have also compared the performance of different spectral analysis methods in distinguishing different conditions. Further, we have examined changes in the EEG spectral characteristics that occur due to migraine in the pregnancy term. ©2010 IEEE.