Skip to main content
. 2023 Jan 24;64(3):692–704. doi: 10.1111/epi.17503

FIGURE 1.

FIGURE 1

Pre‐processing pipeline for normative MEG recordings. First, (A) MEG recordings were pre‐processed to remove any artifacts due to powerline interference, or cardiac and ocular events. After artifact removal, 70 s epochs of channel recordings were source reconstructed and downsampled into cortical regions based on the Lausanne parcellation. Second, (B) regional power spectral densities were computed for each patient. The relative band‐power contribution was calculated for five commonly used frequency bands: delta (1–4 Hz), theta (4–8 Hz), alpha (8–13 Hz), beta (13–30 Hz), and gamma (30–80 Hz). For gamma 47.5–52.5 Hz were discarded to minimize any residual effects due to powerline artifacts. Note that the x‐axis terminates at 50 Hz for illustrative purposes only. Finally, (C) regional band‐power contributions were averaged across the cohort to create normative maps of relative band power for each of the five frequency bands.