Oscillation of the cerebral blood flow (CBF) is a common feature in several physiological or pathophysiological states of the brain. It is promising to identify the disorders of the cerebral circulation based on the classification of CBF signals. In order to distinguish between different physiological states, two different classification methods have been developed; a Radial Basis Function based Neural Network and a Support Vector Classifier with Gaussian kernel. In order to describe the temporal blood flow patterns, two feature extraction procedures were applied; spectral matrix and wavelet subband analysis. These feature extraction and classification methods are evaluated and their efficiencies are compared. The computations were carried out with Mathematica 5.1 and its Wavelet Application. © 2005 IEEE.