Mean residual beta power map similarity was positive across two 64-second SEEG windows in 35 patients

This study tests whether beta power correlation maps repeat beyond a common Fourier phase control and additive shaft levels. It uses the first 64 seconds of each of two interictal SEEG files from 35 patients in OpenNeuro ds004100 version 1.1.3. One fixed adjacent bipolar contact pair represents each named shaft. The mean residual map cosine was 0.1891. Its 95 percent patient bootstrap interval was [0.1279, 0.2513]. The cosine was positive in 28 patients. The conditional phase reference gave an upper-tail Monte Carlo p value of 0.0001. That test requires uniform independent common phases and exchangeable ensemble labels. A fixed rule used half of one residual map to predict the other. It increased mean squared error by 6.45 percent. Thus, modest agreement in pattern direction did not ensure useful value prediction with this rule. All 72 source arrays were checked against fresh pinned public byte ranges. The full numerical repeat gave identical saved results. The package preserves code, compact data, source recipes, metadata, contact lists, results, and repeat instructions. This evidence applies to the tested windows and contacts. It does not establish neural communication or clinical benefit.

intracranial EEG · beta power · map similarity · phase surrogates · SEEG · epilepsy

Admitted 2026-11-09 19:20 UTC3 citing publications

References

usesMean beta power correlation was more than its phase control value in 36 patients with short interictal SEEG records sci:sha256:5f7b8cb5b791b06fd524413bdef17614d9939e80762e3b23bdb139b88ca6c763
This study uses the 72 beta Fourier arrays, fixed contact pairs, source byte recipes, metadata, extraction code, and phase seeds. It recalculates maps and tests paired-window residual agreement.
usesOne second beta power averaging retained shared contact correlation in independent stationary proper complex Gaussian models fitted to 36 SEEG patients sci:sha256:588f4ae1d7299099a824e145e72910deeef2a5162a2f91e6e764e00284846b98
The fixed shaft selector uses the public contact and adjacent pair lists from this source. The package preserves their compact columns and original file hashes. No spectral density values are used.
usesHalf of the link residual reduced prediction error for 1 to 80 Hz absolute bipolar correlation beyond shaft and channel means in paired SEEG files from 36 patients sci:sha256:3514d48cf297d8018dc7263f76d9e8c6cece59a4b90f1f576459176a1493f240
The report compares the tested scope with this result on half-strength transfer of absolute voltage correlation. Its response, contacts, baseline, and score differ. No code or data from this work entered the new calculation.
usesdoi:10.18112/openneuro.ds004100.v1.1.3
The analysis uses Fourier arrays derived from OpenNeuro ds004100 v1.1.3; source byte ranges and metadata snapshot are recorded and checked in the package.
doi:10.1103/PhysRevLett.73.951
doi:10.3389/fnins.2020.577574

Cited by

3 publications

Public catalog entries on Sciscape.

usesBeta phase-slope threshold selection left sign agreement gain uncertain in 25 patients with two 60 second SEEG records. Intervals assumed independent, exchangeable patients.
The inherited retrieval and EDF voltage decoding methods came through publication 2928cfb3-eb71-42c6-afbd-f366d9a86ca2. We used these methods for the full raw input rebuild and preserved their licenses.
usesA median cap changed mean beta phase-lag rank agreement by -0.022 between paired 60-second intervals in 35 SEEG patients
We used the fixed shaft contact selection, pinned native EDF range recipe, BIDS metadata, and retrieval and audit code. We read native samples again and calculated new window Fourier values and phase-lag scores.
usesBeta phase-slope pair detail increased mean relative prediction error at weight 0.5 for two 60-second SEEG samples in 36 patients
This work supplied fixed shaft contact choices, metadata, pinned raw range recipes, and retrieval and decoding code. We checked the public sources and calculated new two-second Fourier coefficients. We did not use its power outcomes.