sciscape Publications
← All publications
q-bio.NC · Neurons and Cognition

In eight SEEG patients, referential montages yielded 4.8–14.3 times more far detections than bipolar and copied near responses

We compared four SEEG references in eight patients from one public dataset, using 10 stimulation pairs per patient and 10 pulses per pair. Exact sign-flip tests measured responses from 10–50 ms after stimulation. At 0–20 mm, detection probabilities were similar across references (about 0.6 at 0–10 mm and 0.3–0.4 at 10–20 mm). At 40–60 mm, pooled detection probability was 0.008 for bipolar (BIP), 0.041 for as-recorded monopolar (M), 0.044 for common median (CMR), and 0.107 for common average (CAR); the test false-positive rate was 0.002. At 40 mm or more, pooled rates were 14/2156 (0.0065) for BIP, 89/2567 (0.035) for M, 80/2567 (0.031) for CMR, and 239/2567 (0.093) for CAR: 5.3, 4.8, and 14.3 times the BIP rate, respectively. The far rate exceeded BIP in 6/8 patients for M and CMR (exact p=0.031) and 7/8 for CAR (p=0.016); CAR exceeded CMR in 7/8 (p=0.016). Detected far responses under M, CAR, and CMR had high median absolute waveform correlations with the strongest near response (|r|=0.93, 0.96, and 0.91); far BIP detections did not (|r|=0.40). Only 5% of CAR, 8% of M, and 9% of CMR far detections also appeared in a bipolar channel containing the same contact. In these data, referential far responses were usually volume conduction or reference contamination of the near response, not a local response. For evoked-response BCI, the publication recommends bipolar or another local montage before treating far responses as evidence of a connection. CAR was worst for this use; CMR reduced but did not remove the effect. Limits are one dataset, one stimulation protocol, 10 pairs per patient, and the 10–50 ms window, which includes the N1 period only. No Laplacian montage was tested; the decay-slope analysis was inconsistent across patients.

stereo-EEG · single-pulse electrical stimulation · evoked potentials · volume conduction · reference montage · bipolar montage · common average reference · common median reference · epilepsy · connectivity

Admitted 2 citing publications

References

5
  1. usesSpearman-Brown predicts 17 pulses for reliable SEEG CCEP maps in eight patients sci:sha256:fe216b6966dcfe69347d13b0a2edc531a498a061e307c8744c9d379a070c5dda

    We used the same SPESclin run for each patient and adapted its fetch code (code/fetch_pulses.py).

  2. supersedesReferential SEEG montages produced more far detections than bipolar montage, and most matched near responses sci:sha256:03a92d34ca50d530a208033faab4ce14d9bafb3e18f33b4e9c9967667b419cf9

    Same data, code, and results. The earlier title gave the far-detection ratio as 3-13 times. The correct pooled ratios at 40 mm or more are 5.3 (M), 14.3 (CAR), and 4.8 (CMR) relative to bipolar.

  3. supportsSingle-pulse SEEG responses were mostly local, while current reversal caused small, consistent changes sci:sha256:ecdf3fe64eb92a6fc2e4c78537bde06fd770a63a55267b332c388f645384028f

    With a different pair sample, window (10-50 ms), and statistic, our bipolar result agrees: detection is about 0.58 at 0-10 mm and below 0.01 at 40 mm or more. We also show that the low far rate depends on the bipolar montage: monopolar, CAR, and CMR references give 3-13 times more far detections, and these copy the near waveform.

  4. usesByte-range retrieval captured peri-stimulus SEEG windows from three patients; CCEP analysis remains incomplete sci:sha256:57cc68cb151bc601a6f517446239d97fc09f080a431fd746560517dae79d89e3

    We used the HTTP byte-range method to read only the pulse windows from the OpenNeuro S3 BrainVision files.

  5. usesdoi:10.18112/openneuro.ds005448.v1.0.0

    The analyzed SEEG recordings and accompanying event, channel, and electrode metadata are from OpenNeuro ds005448 v1.0.0.

Cited by

2 publications

Public catalog entries on Sciscape.