Automatic Identification of Interictal Epileptiform Discharges in Secondary Generalized Epilepsy

Table 2

A full list of features used for the classification of ED segments: denotes all data (source signal) in a segment; denotes data in the nondetected (as a sharp or SW) block only within the segment. Some features are paired (e.g., numbers 9 and 21), which can be calculated for sharps and SWs independently.

Feature number	Description
1	Standard deviation of amplitude for
2, 14	Number of detected waveforms
3, 15	Median wave-width of detected waveforms
4, 16	Median amplitude of detected waveforms
5, 17	Mean kurtosis of detected waveforms
6, 18	Mean ratio of amplitude to wave-width of detected waveforms
7, 19	Standard deviation of amplitude for
8, 20	Feature 8 = feature 1/feature 7; Feature 20 = feature 1/feature 19
9, 21	Feature 9 = feature 4/feature 7; Feature 21 = feature 16/feature 19
10, 22	Max() − min()
11, 23	Max() − min()
12, 24	Feature 12 = feature 4/feature 11; Feature 24 = feature 16/feature 23
13, 25	Mean skewness of detected waveforms
26	Mean distance from a SW to its closest and preceding sharp
27	Ratio of detected waveform types: SW to sharp

Basic features (wave-width, amplitude, kurtosis, amplitude to wave-width ratio, and skewness) were calculated from each detected waveform first, and their median/mean was taken as the features for the classification.