Skip to main content
. 2018 Nov 13;9:889. doi: 10.3389/fneur.2018.00889

Figure 1.

Figure 1

Visual and automated identification of false HFOs. In (A–D), the first rows show one second of iEEG raw data, the second rows show the iEEG signal after the band-pass filtering [80–200 Hz for (A,C) and 200–500 Hz for (B,D)], the third rows show the Morlet wavelet spectrum, while the last rows represent a power spectral density map (PSD). The real HFOs appear as isolated islands (A,C) while false HFOs appear as a mountain-like shape (B,D) in the time-spectrum. On the power spectral density map, true HFOs have obvious peaks in the corresponding frequency range (arrows, A,C), but these peaks do not appear in the false HFOs (B,D).