ABSTRACT
Recognition of spike foci has proved to be a difficult field for EEG automation, and although many methods have been described, progress has been slow. In this chapter, the authors present a review of the major methods used for spike detection. It is only valuable if the threshold values are set for each patient on the basis of previous recordings, thus limiting its use to seizure monitoring. Lieb, Woods, Siccardi, Crandall, Walter, and Leake (1978) used an automatic spike-detection system (COUNTR) based on the second derivative technique. Lopes da Silva et al. and Lopes da Silva, Van Hulten, Lom-men, Storm Van Leeuwen, Van Vellen, and Viegenthart introduced another interesting approach for the detection of epileptic spikes with an automatic spike-detection program. Combinations of several parameters, such as amplitude, duration, and sharpness, maintain the same level of accurate detection, but decrease the number of false positives.
