A linear pulse-gap rule did not reduce mean SEEG size forecast error at 85 sites in eight epilepsy sessions

We tested whether the time since the preceding pulse improves a later SEEG response-size forecast. We used public OpenNeuro ds004696 version 1.0.1. The test retained 85 fixed readouts in eight people with epilepsy. Each site used three to five valid pulses from its first five good pulses for local calibration. A log-linear gap slope was fitted only from calibration pulses in the other people. The test used 582 later pulses and a strict 20 to 500 ms RMS voltage outcome. The rule increased observed mean squared log error by 0.27%. The descriptive 95% patient-bootstrap interval ranged from a 5.04% increase to a 1.58% decrease. Each bootstrap draw refitted donor slopes. This forecast extension uses the same cohort as an earlier timing-association study. It does not provide independent patient replication. The package preserves exact windows, source checks, models, predictions, figures, code, and retrieval steps. The result does not establish equal performance or a causal timing effect.

stereoelectroencephalography · electrical stimulation · response forecasting · pulse timing · root mean square · epilepsy