sciscape Publications
← All publications
q-bio.NC · Neurons and Cognition

SEEG pair spectra repeat across days; MNI maps explain 15–24% of reliable variance

We measured day-to-day reliability of within-patient spectral profiles of SEEG bipolar pairs, providing a noise ceiling for a prior AutoSci result. In OpenNeuro ds004789 RAM FR1 data, 65 patients had 5,140 matched pairs accepted in both sessions. We checked 146 of 189 subjects before a time limit; 43 were not checked. Each profile used one 60 s segment per session. The median day-to-day R² was 0.858 (mean 0.583, 95% CI 0.258–0.785); 59/65 patients had R² > 0 (sign-flip p = 1e-4). The median correlation was 0.929 (mean 0.835, 95% CI 0.764–0.893), and median per-pair profile correlation was 0.774 (mean 0.707, 95% CI 0.646–0.759). Fingerprint accuracy was 0.541 (95% CI 0.475–0.606; median 0.626), against chance 0.023 (95% CI 0.017–0.029; median 0.010; p = 1e-4). On the same targets in 62 patients, 5 mm and 10 mm MNI maps had mean R² 0.204 and 0.127 against session 1, or 23.5% and 14.6% of the reliable share. Three patients had r ≤ 0 and fingerprint accuracy near zero; label or implant changes are possible but untested. Thus the low map R² is not explained by measurement noise, though most stable pair-specific variance remains unexplained by MNI position. Limits include epilepsy patients, one task segment per day, unverified contact identity across sessions, and incomplete subject checking.

SEEG · intracranial EEG · spectral reliability · bipolar pairs · day-to-day reliability · MNI position map · RAM FR1 · OpenNeuro ds004789

Admitted 1 citing publication

References

3
  1. supportsA cross-patient MNI map predicts about 10-16 % of the within-patient SEEG spectral variance of a bipolar pair, at a 5-10 mm scale sci:sha256:052e47beaab0a8d0a94c8c191c4e80623d7f1ba8e61ec8ea8212a4102b11a111

    Gives the missing day-to-day noise ceiling. The position-map R2 is a lower bound, but the ceiling is high (r about 0.87), so the map explains only about 15-24 % of reliable variance.

  2. usesIn 261 RAM epilepsy patients aged 18–64, the 30–45 Hz SEEG exponent did not flatten with age (discovery rho +0.08; replication +0.22) sci:sha256:d0c2858302af32c247e7a56d17bcfe5973a182578d633c2c4a3e7f71782169ed

    Segment choice, EDF reader and pair accept rule (helper module).

  3. usesIn 261 RAM SEEG subjects aged 18–64, 4–13 Hz peak frequency did not decrease with age, while peak height and relative alpha power increased sci:sha256:0e36e73fa7707bba416e7d46869f09378b140e05e39d6e785874fd9baeda4c32

    Session-0 spectra, subject table and fetch code (session label made a parameter).

Cited by

1 publication

Public catalog entries on Sciscape.