Baseline subtraction reduced mean relative SEEG disagreement across all disjoint five pulse splits for selected contacts in five patients

We tested whether scalar baseline subtraction reduced disagreement between separate SEEG pulse means. We used OpenNeuro ds004457 version 1.0.2. The fixed sample had 1081 links, 40 selected contacts, and five patients. Each site had ten pulses. Each comparison used two separate groups of five pulses. The baseline was -120 to -20 ms. The target was 60 to 160 ms. We averaged loss across all 252 oriented splits. We then averaged relative reductions across links and patients. The mean reduction was 46.98%. The descriptive 95% patient bootstrap interval was 30.31% to 66.26%. The exact two-sided patient sign flip p value was 0.0625. All five patient means were positive. The before-stimulation control gave a 48.59% reduction. Scalar subtraction changed mean voltage level and left time-centered shape unchanged. The chronological check depended on how link errors were combined. Lower disagreement does not establish lower neural response error. Every saved voltage bin matched the raw data. The package preserves code, pulse data, source hashes, results, checks, and instructions.

stereoelectroencephalography · SEEG · baseline subtraction · electrical stimulation · pulse averaging · voltage disagreement · signal processing · intracranial EEG