A causal filter with 186 coefficients ended 10 ms gap error effects at 100.6 ms at 2048 Hz when inputs were equal outside the gap
This work tested a finite causal 70 to 110 Hz filter for stereoelectroencephalography (SEEG) readout after stimulation. At 2048 Hz, the filter had 186 coefficients. For two inputs equal outside a 10 ms gap, with equal history and state before the gap, output error was zero from 100.6 ms. The numerical bound on absolute complex gain change was 0.04988. In 400 public epochs from five patients, baseline waveform differences had patient medians of 3.66 to 16.11 percent. The maximum was 126.17 percent. Matched infinite impulse response window bounds were 127.9 to 177.7 ms under a 1000 microvolt root mean square gap error assumption. A later zero DC variant did not improve all patient medians. The package preserves a filter tool, equations, exact digital epochs, four figures, source hashes, and reproduction code. Fresh extraction reproduced all 3,276,800 digital samples. These results concern filter error and offline voltage differences. They do not establish neural recovery or decoder accuracy.
causal filtering · finite impulse response · stimulation artifact · SEEG · brain-computer interface · filter error bound
References
4- usesA 1 mV RMS error model gave 128–178 ms guards for a causal 70–110 Hz filter in five SEEG recordings sci:sha256:2c4619c59bc30f32e3bd9cc14c155e6bd363a5bfae302a15788d0c12105ad785
This work used the causal gap bound code, pinned source manifest, and exact 400 epoch package. It added finite filter support and an all later finite window comparison.
- usesReplacing pulse samples changed zero-phase 70–110 Hz SEEG voltage above baseline RMS at 20–50 ms in 27/400 pulses (five patients). sci:sha256:78299ccd5493ea3f4d9c8ec1dfad43bf77ec6ab49d467ba2f55a3b7b4806dddd
This work used the digital SEEG subset, fixed selection rule, and raw extraction code inherited through the causal bound package. Fresh extraction reproduced the complete subset.
- usesdoi:10.18112/openneuro.ds004457.v1.0.2
The analysis used the archived 400 SEEG epochs selected from OpenNeuro ds004457 version 1.0.2.
- doi:10.1523/JNEUROSCI.1325-22.2023
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesFour polynomial constraints reduced mean pre-stimulation voltage error against a causal 70 to 110 Hz reference in SEEG from 21 patients at 512 and 4096 Hz
This is the original source of the unchanged finite_filter.py code used to calculate the numerical impulse tail bound.
- usesPolynomial constraints reduced baseline error against a causal 70 to 110 Hz reference in 400 SEEG epochs from five patients
This parent supplied the finite impulse and tail code, exact 400 epoch cohort, source manifest, single coefficient DC correction, and offset result. The new work tests a distributed polynomial correction against equal length and parent controls. It repeats the parent filter metrics and source extraction.