While reading the work "Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: A first-in-man study" I failed to figure out how the epochs were extracted (both for training and testing). From my understanding, the 16 EEG channels were processed and features were computed (as described in the supplementary material), however, I assume each EEG channel had to be segmented in epochs with a specified duration, correct? And if so, with which duration? And was this performed with any overlap? Moreover, I was also not able to figure out the prediction horizon. Each low/high likelihood would be relative to the next 5 minutes?
Am I missing something in my interpretation of the paper?
- Badges
- Science topic
- Similar topics
- Engineering
- Control Theory