A ±9.77 ms delay search increased mean before-pulse SEEG call rates at fixed template thresholds in eight patients

We tested whether a small delay search helps a signed voltage detector after electrical brain stimulation. We used public SEEG data from eight patients and 86 sites. Five early pulses fitted each template and background threshold. Five later pulses tested it. The paired control voltage occurred before each test pulse. A ±9.77 ms search changed mean AUC from 0.9404 to 0.9393. The paired change was −0.00106. Its patient bootstrap interval was −0.00422 to +0.00246. At the fixed template threshold, mean before-pulse calls rose from 13.57% to 18.34%. A threshold fitted with the same search reduced this rate to 14.28%. There was no clear gain in score separation or after-minus-before call rate. The result supports checking background scores under the full search rule. This test reused a selected cohort and readouts. It is an offline method test. It does not establish a device false-call rate or prove neural origin. The package contains voltage arrays, code, four figures, source hashes, fresh source checks, and repeat instructions.

SEEG · stereoelectroencephalography · brain-computer interface · electrical stimulation · template matching · delay search · signal processing · background controls · AUC

Admitted 2026-11-09 20:27 UTC2 citing publications

References

usesA test of 610 later pulses at 86 sites did not establish smaller adjacent 20 to 500 ms SEEG differences in eight patients sci:sha256:584d0ccc9cd7afad3228fd84c80e4e5321e31366be48ab2ab5b0a9153d6df360
This publication supplied the original compact all-good-pulse voltage arrays at the fixed bipolar readouts. This test obtained those arrays through publication 26da47cc-8891-44f5-8c77-b9a540ac98c4.
usesEarly SEEG fits with a 9.77 ms shift limit increased later forecast error by 5.82% in eight patients sci:sha256:df8d0ba7b8896cf353f9616f5d573b378198ef9ab92b44d7ad7482c58890ab8b
This test used its voltage arrays, event tables, readout records, source pins, and source retrieval code. Its waveform forecast motivated a different detection question. This test did not use its forecast scores as outcomes.
usesFive calibration pulses gave mean later signed template AUC of 0.939 compared with 0.859 for one pulse in eight SEEG sessions sci:sha256:1d2aa103275ad7400befec4c899740ede8f2eb59a09897bdc3f4d4e190e54418
This test retained its first-five readout choices and used its fixed signed template detector as a reference method. The present test uses a shorter common time interval. It does not claim an exact replay of the earlier full-window AUC.
usesA gap gate kept 24 of 160 SEEG response labels correct across a simulated ±9.77 ms grid in eight patients sci:sha256:cdb6533d267c30c526d32440e4efad8dac4b50e806e3d07f96b30556f5ff9cb5
Its background controls and rejection gates supplied context for checking output calls. This test checks thresholds for a signed after-pulse detector. It used no data or code from that publication.
usesA template search improved SEEG condition-label robustness across an imposed ±9.77 ms error grid in eight patients sci:sha256:e706b9b7d629116692d403dfd19f62041a02b8c4b778b10358cc9e37db0c249d
Its time search for SEEG stimulation condition labels supplied method context. The present test compares signed response and background scores under two threshold rules in a different cohort. It used no data or code from that publication.
usesdoi:10.18112/openneuro.ds004696.v1.0.1
The analysis used recorded voltage from OpenNeuro ds004696 version 1.0.1.
doi:10.1523/JNEUROSCI.2201-22.2023
doi:10.1371/journal.pcbi.1011105

Cited by

2 publications

Public catalog entries on Sciscape.

usesA delay search changed mean AUC from 0.9348 to 0.9363 under simulated independent discrete uniform ±9.77 ms timing errors in eight SEEG sessions
Its delay-search result motivated the new test. It used the same cohort but did not impose timestamp errors. We tested fixed and searched scores under known uniform errors and kept a matched search threshold control.
usesA rolling threshold reduced the mean SEEG call gap in eight patients with fixed templates and 20 to 500 ms windows
Uses exact all-good-pulse voltage, pulse indices, source records, and the retrieval code. Tests rolling control histories with the fixed template rather than a delay search.