Two stage and uniform pulse rules gave 79.1 and 79.4 percent top fifth SEEG site agreement across 7560 allocations in five patients

We tested pulse allocation with public SEEG data from five patients and 152 stimulation-site labels. A two stage rule used two pulses at every site and three more pulses at the highest third of sites. A uniform rule used the same total budget of 463 training pulses. Five separate pulses supplied the test ranking. The score measured repeated voltage structure after a fixed projection and a before-window subtraction. For the first five training pulses, top fifth list agreement was 75.3 percent for the two stage rule and 68.3 percent for uniform sampling. The gain was 7.0 percentage points. Patient resampling limits were -4.2 to 16.2 points; the two-sided sign-flip p value was 0.25. Across all 7560 planned allocations per patient, agreement was 79.1 and 79.4 percent. The allocations reuse ten events and are not independent tests. The average gave no gain from the two stage rule. This result concerns observed list agreement, not known neural connections. The package preserves data, code, scores, source hashes, fresh source checks, and reproduction instructions.

stereoelectroencephalography · electrical stimulation · pulse allocation · site ranking · repeatability · finite-pool analysis