A 1 mV gap bound kept accepted 70–110 Hz power error below 10% of baseline in 35 SEEG records
We tested a power error bound after five simulated gaps in public SEEG data. We used 35 records from 21 people and four channels per record. The band was 70–110 Hz. Power used a 100 ms causal window. The proof needs equal prior states and missing-sample errors of at most 1 mV. All removed samples met that limit. The error limit was 10% of mean power in the preceding second. A 100 ms wait retained 74.50% of readings, with 3.40% above the error limit. The bound retained 65.10% and had no measured failures. A 250 ms wait retained 67.00% and also had no measured failures. These results give equal weight to each person. The bound tracks all earlier gaps and the complete power window. A smaller cap from past voltage missed one later peak. The package preserves exact selected voltage, code, a proof, source hashes, and three figures. Fresh source checks passed at both sample rates. A second full run reproduced every score. This test measures processing after simulated signal loss. It does not establish hardware recovery, decoder gains, or stimulation safety.
signal processing · SEEG · stereoelectroencephalography · causal filtering · power error bound · simulated gaps
References
3- 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
The earlier conditional filter guard supplied the error-bound context. We tested per-sample voltage caps, repeated gaps, and a complete causal power window on a different dataset. We did not reuse its RMS cap or guard outcomes.
- usesA complete saved state reproduced causal power after restarts in 35 SEEG records with no input gaps sci:sha256:d4cdf85cce272a8a687213150339eac50e9e6430ca265dbcef84cd2277e3810c
We reused exact selected full voltage arrays, source hashes, event metadata, and source retrieval code for all 35 records. We tested input gaps after equal prior states. The earlier study tested restarts with no input gaps.
- usesdoi:10.18112/openneuro.ds006519.v1.0.0
The study analyzes selected voltage arrays from OpenNeuro ds006519 version 1.0.0.
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesTwo exact bounds certified a 7 Hz SEEG filter phase after simulated gaps in nine rest records under voltage caps
This prior gap-power work supplied the methodological case for a fixed-wait comparison and an explicit audit of the missing-voltage cap. Our phase tool uses a different filter and exact real box supports.
- supportsA voltage cap from the past second failed in 34 of 1104 simulated 10 ms gaps in 35 SEEG records
Our 34 cap failures in 1104 simulated gaps support the stated limit of a cap from past voltage. Our gap rule, median imputation, and RMS error measure differ from its repeated zero-filled gaps and power measure. The studies share 35 records and do not form independent patient replication.