Abstracts

Rationale: The NeuroPace RNS System provides an unprecedented clinical tool for recording brain activity around patients' epileptic foci. However, its use in research is complicated by the difficulty of synchronizing RNS recordings with external time sources. While our previous work used a wand accessory to insert highly precise synchronization markers, these methods are impractical for use in ambulatory experiments. The aim of this study is to present a novel method for synchronizing RNS recordings accurately to task events while preserving the capability to acquire ambulatory recordings.  Methods: We used a magnetometer mounted on a subject-worn headset to continuously measure the magnetic field strength near the neurostimulator. The subjects were instructed to swipe a small magnet across their neurostimulator twice in opposite directions. The times at which the neurostimulator detected the magnet swipes and the continuous recordings from the magnetometer were used to determine a clock offset between the two systems.  Results: We found that the midpoint of two magnet swipes closely aligned to the RNS's onset of recording with reliable consistency. From preliminary trials (n = 77), the mean time difference between the magnetometer system and RNS was -0.204 seconds with a variance of 0.403 seconds.  Conclusions: Synchronization using magnetometer readings was able to reduce the contribution to timing error due to poor synchronization so that it is no longer the dominating source of timing error in magnet-triggered recordings. Unlike the synchronization marker technique, this technique is compatible with ambulatory recordings, expanding the RNS's application into nonstationary studies.  Funding: Diamond Foundation Research Development Award