Adding a fixed response mean score did not improve average SEEG site accuracy at 20–100 ms in five offline patient tests with four calibration pulses per site

We tested a fixed SEEG site reader with public data from five patients in OpenNeuro ds004457 version 1.0.2. Each site had four calibration pulses and four separate test pulses. The primary response window was 20 to 100 ms. A second reader added a response mean measured from a pre-pulse baseline. Average patient accuracy changed from 40.56% to 39.89%. The change was -0.66 percentage points. Its descriptive 95% patient bootstrap interval was -1.69 to +0.83 points. Four patients had lower accuracy. This test did not establish a benefit or harm from the added mean. Calibration and test times overlapped across sites. This was a retrospective test in reused clinical sessions. The package preserves compact input data, all scores, code, source hashes, three figures, and reproduction instructions. A fresh raw extraction reproduced all ten input files byte for byte. A second analysis reproduced every numerical output. This study did not test a live BCI or a neural write effect.

stereoelectroencephalography · brain-computer interfaces · electrical stimulation · response centering · site classification · calibration · baseline voltage