Analysis-rcs-data: Open-Source Toolbox for the Ingestion, Time-Alignment, and Visualization of Sense and Stimulation Data From the Medtronic Summit RC+S System

Abstract

Closed-loop neurostimulation is a promising therapy being tested and clinically implemented in a growing number of neurological and psychiatric indications. This therapy is enabled by chronically implanted, bidirectional devices including the Medtronic Summit RC+S system. In order to successfully optimize therapy for patients implanted with these devices, analyses must be conducted offline on the recorded neural data, in order to inform optimal sense and stimulation parameters. The file format, volume, and complexity of raw data from these devices necessitate conversion, parsing, and time reconstruction ahead of time-frequency analyses and modeling common to standard neuroscientific analyses. Here, we provide an open-source toolbox written in Matlab which takes raw files from the Summit RC+S and transforms these data into a standardized format amenable to conventional analyses. Furthermore, we provide a plotting tool which can aid in the visualization of multiple data streams and sense, stimulation, and therapy settings. Finally, we describe an analysis module which replicates RC+S on-board power computations, a functionality which can accelerate biomarker discovery. This toolbox aims to accelerate the research and clinical advances made possible by longitudinal neural recordings and adaptive neurostimulation in people with neurological and psychiatric illnesses.

Decoding natural gait cycle in Parkinson’s disease from cortico-subthalamic field potentials

“…We used wavelet transformation because it has greater low-frequency resolution, which is of interests given that our data showed low frequency power modulations during gait. We also used the Fourier transform because this is the spectral decomposition method used by the Summit RC+S system (Sellers et al, 2021).…”

“…While current continuous DBS (cDBS) can improve many motor symptoms of PD, it is often ineffective to alleviate gait disturbances given that it does not account for the rapid dynamic changes that occur with the gait cycle (Wang and Choi, 2020). The Summit RC+S system implanted in our subjects allow for aDBS in real time and utilizes LDA to detect different brain states using Fourier transform power within a frequency band (Ansó et al, 2022;Sellers et al, 2021). The aDBS feature of the Summit RC+S device has been successfully tested in single cervical dystonia patient (Johnson et al, 2021), and in PD patients (Gilron et al, 2021a(Gilron et al, , 2021b.…”

The Role of Large-Scale Data Infrastructure in Developing Next-Generation Deep Brain Stimulation Therapies

“…and stimulation settings (frequency, amplitude, ramp rates, etc.) ( Sellers et al, 2021 ). Importantly, full integration into both patient and researcher workflows ensures that users need not be involved in the engineering processes that handles data transfer, upload, and parsing ( Figure 1B ).…”

PELP: accounting for missing data in neural time series by Periodic Estimation of Lost Packets

“…As an exemplar device, we focus our methodological development on the wireless data transmission from the Summit RC+S, however, the methodology is generally applicable. For recordings using the Summit RC+S, each packet has three integer timing variables of note for this purpose: ‘dataTypeSequence’ indicating the packet number that rolls over every 256 packets, ‘systemTick’ time of the last sample in a packet with 0.1 ms resolution that rolls over every 6.5536 s, and ‘timestamp’ with 1 s resolution and no rollover [22]. A packet loss has occurred when the dataTypeSequence between subsequent packets skips an index or the timestamps are inconsistent with the systemTick data.…”