Beschreibung:
<jats:title>Summary</jats:title><jats:sec><jats:title>Objective</jats:title><jats:p>We investigated the performance of automatic spike detection and subsequent electroencephalogram (<jats:styled-content style="fixed-case">EEG)</jats:styled-content> source imaging to localize the epileptogenic zone (<jats:styled-content style="fixed-case">EZ</jats:styled-content>) from long‐term <jats:styled-content style="fixed-case">EEG</jats:styled-content> recorded during video‐<jats:styled-content style="fixed-case">EEG</jats:styled-content> monitoring.</jats:p></jats:sec><jats:sec><jats:title>Methods</jats:title><jats:p>In 32 patients, spikes were automatically detected in the <jats:styled-content style="fixed-case">EEG</jats:styled-content> and clustered according to their morphology. The two spike clusters with most single events in each patient were averaged and localized in the brain at the half‐rising time and peak of the spike using <jats:styled-content style="fixed-case">EEG</jats:styled-content> source imaging. On the basis of the distance from the sources to the resection and the known patient outcome after surgery, the performance of the automated <jats:styled-content style="fixed-case">EEG</jats:styled-content> analysis to localize the <jats:styled-content style="fixed-case">EZ</jats:styled-content> was quantified.</jats:p></jats:sec><jats:sec><jats:title>Results</jats:title><jats:p>In 28 out of the 32 patients, the automatically detected spike clusters corresponded with the reported interictal findings. The median distance to the resection in patients with Engel class I outcome was 6.5 and 15 mm for spike cluster 1 and 27 and 26 mm for cluster 2, at the peak and the half‐rising time of the spike, respectively. Spike occurrence (cluster 1 vs. cluster 2) and spike timing (peak vs. half‐rising) significantly influenced the distance to the resection (p < 0.05). For patients with Engel class <jats:styled-content style="fixed-case">II</jats:styled-content>,<jats:styled-content style="fixed-case"> III</jats:styled-content>, and <jats:styled-content style="fixed-case">IV</jats:styled-content> outcomes, the median distance increased to 36 and 36 mm for cluster 1. Localizing spike cluster 1 at the peak resulted in a sensitivity of 70% and specificity of 100%, positive prediction value (<jats:styled-content style="fixed-case">PPV</jats:styled-content>) of 100%, and negative predictive value (<jats:styled-content style="fixed-case">NPV</jats:styled-content>) of 53%. Including the results of spike cluster 2 led to an increased sensitivity of 79% <jats:styled-content style="fixed-case">NPV</jats:styled-content> of 55% and diagnostic <jats:styled-content style="fixed-case">OR</jats:styled-content> of 11.4, while the specificity dropped to 75% and the <jats:styled-content style="fixed-case">PPV</jats:styled-content> to 90%.</jats:p></jats:sec><jats:sec><jats:title>Significance</jats:title><jats:p>We showed that automated analysis of long‐term <jats:styled-content style="fixed-case">EEG</jats:styled-content> recordings results in a high sensitivity and specificity to localize the epileptogenic focus.</jats:p></jats:sec>