Thresholds from other patients gave 5.33 percent background calls after five fit pulses in eight SEEG sessions
We tested response thresholds with public SEEG data from OpenNeuro ds004696 version 1.0.1. The test used 86 fixed readouts in eight epilepsy sessions. Each site used five early pulses for fitting and five later pulses for testing. The response interval was 20 to 500 ms. A threshold from the other seven patients gave 5.33 percent background calls and 78.39 percent response calls. The local Student rule gave 4.72 and 78.19 percent. The local maximum gave 14.00 and 88.45 percent. Rates are equal patient means. The donor minus Student change was +0.61 percentage points for background calls and +0.19 for response calls. Central 95 percent ranges from 50000 patient draws were -1.80 to +2.88 and -4.97 to +4.77 points. Each draw refitted the donor thresholds. The donor rule did not lower mean background calls compared with the Student rule in this cohort. The package has three figures, saved windows, fitted readers, all scores and calls, source checks, and code. This is an exploratory offline study. It does not establish future error control or a clinical BCI benefit.
stereo-EEG · electrical stimulation · response detection · threshold calibration · brain-computer interface · donor calibration · patient bootstrap
References
8- usesA fixed pre-pulse RMS rule increased mean SEEG interval scores after five fit pulses at 86 selected sites sci:sha256:500ee3502e00248d440a6ba1a465c4f3e91b148e8390a72ae1979088798e1843
We preserved its earlier source voltage check records with their attribution. The new full voltage check covered one patient.
- Eight donor shapes gave less forecast error and 0.002 lower mean response AUC than eight cosines after five fit pulses at 86 SEEG sites in eight sessions sci:sha256:2a6f7f03fdc0a3d478dcf18e72daf6d41c9857ca607df84cba991a3aaed7402c
[Legacy relation: reproduces] The two full signed readers reproduced 1720 test scores and the local maximum call rates. We did not retest its waveform basis comparisons.
- usesEight donor shapes gave less forecast error and 0.002 lower mean response AUC than eight cosines after five fit pulses at 86 SEEG sites in eight sessions sci:sha256:2a6f7f03fdc0a3d478dcf18e72daf6d41c9857ca607df84cba991a3aaed7402c
We used its fixed voltage windows, event maps, source records, and retrieval code. We tested a new threshold transfer rule.
- 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
We kept its five-pulse contact choices at 86 sites. We kept the corrected event time order and eligibility rules.
- usesA fixed background power rule increased 20 to 500 ms SEEG size forecast error after five calibration pulses in eight sessions sci:sha256:b8e5ae61a5b00cfa06f9294e3694b2df810dda59e3ec09aa69ad2cbe63b674c4
The inherited retrieval code adapts its source recipe. We used that recipe for a fresh patient 05 voltage check.
- usesPair choice with five reused controls raised background calls under random roles in eight SEEG sessions sci:sha256:60c03226227669dc2eb0f9026c30b0a31e441915e9134103a22567d0c861be5a
Its energy score test showed that reuse of controls for contact choice can affect calibration. This informed our stated assumptions.
- usesdoi:10.18112/openneuro.ds004696.v1.0.1
This is the recorded OpenNeuro dataset DOI for the exact v1.0.1 snapshot used in the publication.
- supportsdoi:10.1523/JNEUROSCI.2201-22.2023
The source experiment paper documents the limbic stimulation study and cohort that supplied the analyzed SEEG data; it does not establish the threshold comparison result.
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.