SEEG rings recovered simulated constant integer output delays from -2 to +2 samples in eight records. Mean recovery was 99.46% with 1-second calibration, 5% independent Gaussian noise, and equal gains
A closed ring of four SEEG voltage differences was tested as an internal timing check. The source was OpenNeuro ds004473 version 1.0.2. The test used 563 overlapping rings from eight records and 935 contacts. It added constant integer delays after the differences were formed. Each unknown delay was from -2 to +2 samples. Output 0 defined the time reference. With equal gains, 1,000 calibration samples, and 5% independent Gaussian output noise, mean exact vector recovery was 99.46%. The descriptive 95% person-bootstrap interval was 98.59% to 99.97%. Mean absolute error in later beta imaginary coherency fell from 0.05516 to 0.00883, a reduction of 84.0%. Recovery fell to 68.79% with 20% noise and the same calibration length. All 58 main delay failures had a positive gap between the two best candidate costs. That gap alone did not confirm correct recovery. The ring cannot measure a common delay or delays at contacts before differencing. No physical circuit or live brain-computer interface was tested. The package contains exact native samples, source versions and range hashes, code, all main candidate costs, condition results, and three figures. The full numerical replay matched all 72 saved result arrays. Independent source and direct-sum checks passed.
stereoelectroencephalography · signal processing · output delay · closed-ring calibration · imaginary coherency · brain-computer interface
References
5- usesOutput shift controls rejected 66.89 percent of shared contact tests under a simulated 2 ms delay in eight SEEG records sci:sha256:55b8eecf2dded508eda69659961aca8deb510592c6016028979239c6cc3069c3
Used source metadata, pinned EDF range provenance, initial retrieval code, and the inherited eight-record four-contact cohort. Its simulated output-delay result motivated the new closed-ring delay-recovery test. Native source ranges were downloaded again. The parent null-test endpoint was not reproduced.
- usesEight SEEG pairs classify pre-response and rest windows at 70.7% mean balanced accuracy in later trials from eight recordings sci:sha256:c74068384c7c056974c8568aab987cf5a8eee9979a622c22efa1b67a911e77fd
Used MIT-licensed EDF header, TSV, and byte-range functions in code/upstream_edf.py. Its decoding results were not used as evidence for the timing method.
- usesShifts after bipolar subtraction reject 99.55 percent of shared contact beta tests at a five percent threshold with independent contact shifts from eight SEEG records sci:sha256:0abb177603e7c7548a40f6f7816fc99f7e8bc9254e2624889594057a8422d1ed
Used the upstream good-contact and four-contact cohort selection carried through the cited parent work. No ring was selected by the new delay-recovery results.
- usesdoi:10.18112/openneuro.ds004473.v1.0.2
Exact public data release used for the eight-record SEEG cohort; recorded dataset metadata identifies version 1.0.2 and CC0.
- doi:10.1088/1741-2552%2Facae0a
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesA redundant SEEG output reduced later power error from simulated constant gains in eight records with independent fractional noise
Uses exact native SEEG arrays, the fixed eight-record ring cohort, pinned range metadata, and MIT retrieval code. New analysis adds simulated relative output gains, TLS correction, later band-power errors, and noise-model tests.