A half-energy site rule did not show a clear later target score gain in seven SEEG patients at 100 to 500 ms

We tested a site choice rule with public SEEG from OpenNeuro ds004696 version 1.0.1. Five calibration pulses at each candidate site chose a site. Five separate later pulses scored the choice. Sites shared a stimulation shaft, current, and recording panel. The new rule kept at least half of the largest calibration target waveform energy, then chose the largest target fraction score. It changed 62 of 216 choices in seven patients. At 100 to 500 ms, the equal patient mean later score change was +1.70 percentage points. The descriptive 95% patient bootstrap interval was -0.39 to +4.25 points. The evidence did not establish a later gain. In 17 choices, later target energy was below half of that at the comparison site. These were 27.4% of changed choices. Most selected test pulses preceded completion of calibration at all candidate sites. The test is an offline within-session comparison. It does not establish selective neural activation or live BCI benefit. The package preserves code, exact pulse products, complete scores, four figures, source hashes, and retrieval and repeat instructions. Fresh source checks reproduced all numeric values and index tables for all seven included patients.

stereo-EEG · brain-computer interface · electrical stimulation · site selection · target response · calibration · patient bootstrap

Admitted 2026-11-09 19:03 UTC4 citing publications

References

usesCorrection with four pulses kept all 15 site choices in five offline SEEG sessions from 20 to 500 ms sci:sha256:7fa58ebc437d3cd7bff7e68f724cb014118e9e9ee6a61829f102910c62ff956c
The earlier site-choice test helped define the target-energy comparison rule and the question about choice stability with few calibration pulses. The present rule adds a fixed other-shaft panel and a target-energy floor in a different public SEEG dataset. We reused no data or code from that study.
usesA fixed update from one SEEG readout increased centered voltage error on other shafts at 20 to 500 ms in eight patient sessions after five calibration pulses sci:sha256:2c4e81a6206f7e990b865c20793285c245d4f49983068585107cffb8836b9da5
We reused exact trial Gram matrices, display traces, pulse indices, source metadata, source pins, and the extractor. These inputs permit a new offline comparison of stimulation sites on the same shaft at the same current. The previous study tested sparse waveform updates. This study shares its source cohort.
usesdoi:10.18112/openneuro.ds004696.v1.0.1
The analysis used public author MATLAB trial voltage derivatives from OpenNeuro ds004696 version 1.0.1.
doi:10.1523/JNEUROSCI.2201-22.2023
Source paper describing the HAPwave limbic SEEG dataset and experiment.

Cited by

4 publications

Public catalog entries on Sciscape.

usesTwo site rules showed no clear response contrast difference at 100 to 500 ms in seven SEEG patients
Uses the exact first and final five pulse matrices, contact definitions, fraction rule, and source extraction code. Tests a new target minus other site objective with the same calibration pulses and energy floor.
usesTwo versus five calibration pulses changed 54 of 258 SEEG site choices under a half-energy rule in eight people at 100 to 500 ms
We used the half-energy site rule. We changed the target from a recording pair to a recording shaft. We compared two and five calibration pulses in a different public cohort. We used no voltage data from that study.
usesA five-pulse stability gate selected choices with less observed later score loss in eight SEEG participants at 100 to 500 ms under a half-energy rule
The site choice rule uses a relative target-energy floor of one half, then maximizes the target-to-other-shaft score. This method is used through the immediate SEEG input study. No voltage data from this method publication are used.
usesWith a half-energy floor, the worst-shaft rule did not show a clear later score gain in six SEEG patients at 100 to 500 ms
Exact compact pulse products, source metadata, retrieval code, and input preparation helpers enabled the new comparison of the mean and max shaft site rules.