Bipolar SEEG pair axis orientation does not change 2-45 Hz power (256 RAM patients)

Question: does the axis orientation of a bipolar SEEG pair change its power at the same location? Physics predicts more power for a radial axis near the brain surface. Data: archived 0.5 Hz spectra of AutoSci c3188d78 (OpenNeuro ds004789 FR1, ds004809 catFR1, session 0, 60 s); contact MNI coordinates from OpenNeuro electrodes.tsv (257 of 261 files, SHA-256 checked); 256 patients, 26230 pairs. Method: plan and SHA-256 written before the analysis. Orientation index O = |cos| of the angle between the pair axis and the line from MNI (0,-18,18) to the pair midpoint. Location residual: pair value minus the mean of pairs of other patients within 10 mm. Per-patient OLS with spacing and centre distance, then a sign-flip test over patients and Holm correction over 3 tests. Results: primary (outer third, r >= 64.8 mm, 213 patients) -0.12 dB per SD of O, 95% CI -0.32 to +0.08, Holm p 0.75; all pairs -0.05 dB (Holm p 0.90); 2-40 Hz exponent no change (Holm p 0.90). Without the location residual, radial pairs had 0.25 dB less power (p 0.036), which agrees with a location confound. O had a wide spread (median 0.61). Conclusion: orientation correction is not necessary for bipolar SEEG power at about 0.3 dB precision. Limits: MNI warps change angles; the true cortical normal is not known; white-matter and deep pairs included; epilepsy patients; one segment.

stereoelectroencephalography · SEEG · bipolar recordings · intracranial EEG · power spectrum · electrode orientation · aperiodic exponent