The sensitivity of ECG contamination to surgical implantation site in adaptive neurostimulation

Abstract

ObjectiveInvestigate neural signal contamination with electrocardiographic (ECG) activity in bidirectional implantable neurostimulation systems for closed-loop stimulation in neurological diseases.MethodsElectric field modelling was used to predict the relevance of implant location to ECG contamination in neural signals recorded with implantable devices. Signals from 335 recording streams in 136 hemispheres from 85 patients were visually inspected for ECG contamination to validate the model. Implant sites were categorized into left, right subclavicular hemibody and cranial implant.ResultsOur model predicted significant dependence of the ECG amplitude on implant location, ranging from a 4x difference with chest location to a 1000x suppression with cranial implants. The impact on neural recordings depends on the difference between ECG signal and noise floor of the recording. Empirically, we demonstrate that severe ECG contamination was more than 3.2x higher in left implants (48.3%), when compared to right implants (15.3%). Cranial implants did not show ECG contamination.ConclusionGiven the relative frequency of corrupted neural signals, we conclude that implant location will have significant impact on patient recruitment and feasibility in multicenter clinical trials for adaptive brain stimulation. Prospectively, the impact of implant location on signal fidelity should be carefully considered in the surgical planning process.

Chronic Sensing of Subthalamic Local Field Potentials: Comparison of First and Second Generation Implantable Bidirectional Systems Within a Single Subject

“…Other sources of noise previously reported with the PC + S, arising from interactions between sampling clocks and stimulation rates, were not seen with the Percept PC ( Goyal et al, 2021 ). ECG has remained a persistent source of artifact in many recordings from Percept PC (affecting 65.2% of left sub-clavicular Percept PC implants in one report, Neumann et al, 2021 ), though this was not seen in our patient. LFPs collected from the Percept in this patient did, however, demonstrate a repetitive artifact ( Figure 3C ) not previously described with either device.…”

“…patient among those implanted with the Percept PC at our institution and in the available literature (Goyal et al, 2021;Neumann et al, 2021). This artifact was apparent both from data at postoperative day 0 (in the post-anesthesia care unit) and postoperative day 9 (in the movement disorders clinic), excluding an environmental source.…”

Machine Learning Will Extend the Clinical Utility of Adaptive Deep Brain Stimulation

“…The difference from He et al likely lies in the higher noise floor of the implant when compared to the dedicated laboratory equipment available in experimental settings with externalized leads. Next‐generation sensing‐enabled implantable DBS devices already demonstrate higher signal fidelity but can still suffer from other sources of artifacts, such as cable movement or electrocardiographic activity 10 …”

“…Instead, they followed a strategy to decode the presence of tremor-provoking voluntary movement or postures using a combination of 12 different signal features to adapt stimulation before tremor emerged. Features comprised frequency domain band power measures (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37)(38)(39)(40)(41)(42)(43)(44)(45) and statistical properties of timedomain signals (mean, complexity, mobility, and activity). The decoding strategy was supported by offline model comparisons, where the predictive value of each approach is assessed by splitting recorded data into a training and test set.…”

Musical Hallucinations in Chronic Pain: The Anterior Cingulate Cortex Regulates Internally Generated Percepts

“…In this case, the recordings collected during right ACC stimulation were excluded due to stimulation artifact. Further, neural recordings from the left ACC were excluded from the results due to confounding ECG artifact in the power bands of interest, as previously (84)(85)(86). In these studies, authors report on the current leakage of cardiac activity into the IPG due to the proximity of the left pectoral implant to the heart and recommend right IPG sensing for more cleaner recordings (at 3.2× less contamination than left pectoral implants) (84,85).…”