Three SEEG recording pairs share channel hashes and differ in event labels in ds004457 v1.0.2 and ds004696 v1.0.1
We tested exact recording reuse in three fixed public SEEG releases. The audit includes 4,079 raw channel objects from 19 recordings. Three recording pairs in ds004457 v1.0.2 and ds004696 v1.0.1 have equal complete sets of named channel keys. These pairs include 498 SEEG channel objects. We downloaded six selected complete channel files and checked all 75,478,624 bytes. All six hashes match their source keys. The linked recordings have four changed pulse-time labels, 32 changed quality labels, and 71 changed contact-number texts. The largest time change is 562.988 ms. For this event, two 20–500 ms response windows share no samples. We supply a reuse map, a data split check, voltage examples, three figures, and reproduction code. The identity result assumes correct source keys and no SHA-256 collision. Different hashes do not exclude edited copies. The work does not measure a change in BCI accuracy or test stimulation safety.
stereoelectroencephalography · data reuse · content hashes · event labels · brain-computer interfaces · dataset leakage · electrical stimulation
References
9- usesA fixed covariance rule gave higher SEEG site accuracy than a diagonal rule in offline tests from five patient sessions sci:sha256:1407cd9ac712a4bcd551a3738cc90c3fab309b871572cbd6fc644b41555f7886
The summary of this within-session site-decoding test motivated the BCI use of ds004457 pulse-response data. We did not copy its code or model outputs. We make no claim that it used the two linked releases as independent data.
- usesA five-pulse check selected more repeatable bipolar SEEG pairs within eight epilepsy patient sessions sci:sha256:b5b7963746f5c6d2a2a8f7844ec590732a4eb152bc7a70979b3bc95c3e27820f
The summary of this within-session repeatability check motivated the BCI use of ds004696 pulse-response data. We did not copy its code or derived voltages. We make no claim that it used the two linked releases as independent data.
- usesds003848 v1.0.3 and ds005448 v1.0.0 share one raw SEEG recording sci:sha256:bc1dc109c6929376c3d6b21ee71e676faf097e5772b56d7e5796835e340c7e17
This earlier report motivated the use of content hashes and alias groups to check raw SEEG reuse. The present audit tests different MEF releases and adds event-label and voltage checks. All new analysis code was written for this package.
- usesdoi:10.18112/openneuro.ds004457.v1.0.2
The audit uses this exact OpenNeuro release; source metadata and fixed file hashes are preserved.
- usesdoi:10.18112/openneuro.ds004696.v1.0.1
The audit uses this exact OpenNeuro release; source metadata and fixed file hashes are preserved.
- usesdoi:10.18112/openneuro.ds004977.v1.2.0
The audit uses this exact OpenNeuro release; source metadata and fixed file hashes are preserved.
- doi:10.1523/JNEUROSCI.1325-22.2023
- doi:10.1523/JNEUROSCI.2201-22.2023
- doi:10.1016/j.jneumeth.2024.110153
Cited by
6 publicationsPublic catalog entries on Sciscape.
- usesA fixed noise correction reduced later 20 to 500 ms waveform error at SEEG pairs selected from metadata in eight recordings
Used the recording hash audit to identify reused sources and to limit claims about independent evidence.
- usesA limited time trend increased SEEG forecast error by 9.6% against the past mean in five 6 mA sessions at 20 to 100 ms
Its raw recording reuse map shows why this study and the prior ds004696 trend study do not provide independent evidence from 13 patients.
- usesFive-pulse SEEG selection with a before-pulse score gave no clear 20 to 500 ms test gain in eight people
The recording-overlap report supplied evidence for the cohort reuse limit. We use only ds004696 and do not count overlapping releases as independent cohorts. We did not repeat that audit.
- usesAt selected pairs in eight SEEG recordings, scoring all later pulses changed normalized 20 to 500 ms error of a fixed five-pulse forecast by less than 0.7% per person
We used the channel key inventory and MEF access example to locate public source files. We checked each selected key against the pinned source tree.
- usesFinal five pulses showed no clear 20–100 ms SEEG order contrast at fixed readouts across 314 sites in eight people
Its complete channel-key audit identifies three recordings shared by ds004457 and ds004696. We use that map to state the overlap with prior pulse-order evidence. We do not count these releases as independent cohorts or repeat the full recording-reuse audit.
- usesThe mean response minus pre-pulse SEEG order score was 0.001 at 15 fixed readouts in five sessions at 20 to 500 ms
This audit found three shared recordings in ds004457 and ds004696. That evidence prevents an independent-cohort interpretation of agreement with the ds004696 pulse-order audit. The report states this limit.