Precisely-timed outpatient recordings of subcortical local field potentials from wireless streaming-capable deep-brain stimulators: a method and toolbox.

Background: Investigations of the electrophysiological mechanisms of the human subcortex have relied on recording local field potentials (LFPs) during deep-brain stimulation (DBS) neurosurgery. However, the neurosurgical setting severely restricts the research use of these recordings. Recently developed sensing-capable DBS devices wirelessly stream subcortical LFPs in outpatient settings. These recordings have tremendous potential for research. However, synchronizing them with other behavior or neural recordings is challenging, as the clinical devices do not accept digital timing information.

Methods: Switching the DBS device on introduces transient yet consistent artifacts in both the LFP and simultaneous scalp-EEG recordings. We use these artifacts as a reference to align these recordings (N=20). We tested whether the alignment was precise enough to match a ground truth state (large artifacts produced by transcranial magnetic stimulation, TMS), yielded trial-averaged event-locked LFPs, and phase consistency across trials. We further evaluated the consistency of task-related LFPs across outpatient and perisurgical recordings. Previous alignment methods were limited because they relied on inconsistent on/offset features of DBS artifacts caused by ongoing stimulation. Moreover, they only provided limited validation. Our highly precise alignment method showed a maximum deviation of only 8 ms - clearly superior to prior techniques. Furthermore, event-related activity patterns were comparable across outpatient and perisurgical LFP recordings.

Conclusions: We present a method and a MATLAB toolbox that inserts the most precise digital timing information into wirelessly-streamed DBS-LFP recordings to date. By enabling event-related research with high-temporal precision, this method greatly enhances the utility of these recordings.