Automatic spike detection via an artificial neural network using raw EEG data: effects of data preparation and implications in the limitations of online recognition

Objective: Automatic detection of epileptic EEG spikes via an artificial ne ural network has been reported to be feasible using raw EEG data as input. This study re-investigated its suitability by further exploring the effects of data preparation on classification performance testing. Methods: Six hundred EEG files (300 spikes and 300 non-spikes) taken from 2 0 patients were included in this study. Raw EEG data were sent to the neura l network using the architecture reported to give best performance (30 inpu t-layer and 6 hidden-layer neurons). Results: Significantly larger weighting of the 10th input-layer neuron was found after training with prepared raw EEG data. The classification process was thus dominated by the peak location. Subsequent analysis showed that o nline spike detection with an erroneously trained network yielded an area l ess than 0.5 under the receiver-operating-characteristic curve, and hence p erformed inferiorly to random assignments. Networks trained and tested usin g the same unprepared EEG data achieved no better than about 87% true class ification rate at equal sensitivity and specificity. Conclusions: The high true classification rate reported previously is belie ved to be an artifact arising from erroneous data preparation and off-line validation. Spike detection using raw EEG data as input is unlikely to be f easible under current computer technology. (C) 2000 Elsevier Science Irelan d Ltd. All rights reserved.