Lim Jeffrey, Wang Po T, Shaw Susan J, Gong Hui, Armacost Michelle, Liu Charles Y, Do An H, Heydari Payam, Nenadic Zoran
Department of Biomedical Engineering, University of California, Irvine, Irvine, CA, United States.
Rancho Los Amigos National Rehabilitation Center, Downey, CA, United States.
Front Neurosci. 2022 Oct 12;16:1021097. doi: 10.3389/fnins.2022.1021097. eCollection 2022.
Cortical stimulation electrocorticography (ECoG) may be an effective method for inducing artificial sensation in bi-directional brain-computer interfaces (BD-BCIs). However, strong electrical artifacts caused by electrostimulation may significantly degrade or obscure neural information. A detailed understanding of stimulation artifact propagation through relevant tissues may improve existing artifact suppression techniques or inspire the development of novel artifact mitigation strategies. Our work thus seeks to comprehensively characterize and model the propagation of artifacts in subdural ECoG stimulation. To this end, we collected and analyzed data from eloquent cortex mapping procedures of four subjects with epilepsy who were implanted with subdural ECoG electrodes. From this data, we observed that artifacts exhibited phase-locking and ratcheting characteristics in the time domain across all subjects. In the frequency domain, stimulation caused broadband power increases, as well as power bursts at the fundamental stimulation frequency and its super-harmonics. The spatial distribution of artifacts followed the potential distribution of an electric dipole with a median goodness-of-fit of = 0.80 across all subjects and stimulation channels. Artifacts as large as ±1,100 μV appeared anywhere from 4.43 to 38.34 mm from the stimulation channel. These temporal, spectral and spatial characteristics can be utilized to improve existing artifact suppression techniques, inspire new strategies for artifact mitigation, and aid in the development of novel cortical stimulation protocols. Taken together, these findings deepen our understanding of cortical electrostimulation and provide critical design specifications for future BD-BCI systems.
皮层刺激脑电图(ECoG)可能是在双向脑机接口(BD-BCI)中诱导人工感觉的有效方法。然而,电刺激引起的强烈电伪迹可能会显著降低或掩盖神经信息。详细了解刺激伪迹在相关组织中的传播可能会改进现有的伪迹抑制技术,或激发新型伪迹减轻策略的发展。因此,我们的工作旨在全面表征和建模硬膜下ECoG刺激中伪迹的传播。为此,我们收集并分析了四名植入硬膜下ECoG电极的癫痫患者在明确皮层映射过程中的数据。从这些数据中,我们观察到所有受试者的伪迹在时域中均表现出锁相和棘轮特性。在频域中,刺激导致宽带功率增加,以及在基本刺激频率及其超谐波处的功率突发。伪迹的空间分布遵循电偶极子的电位分布,所有受试者和刺激通道的中位拟合优度为 = 0.80。高达±1100 μV的伪迹出现在距刺激通道4.43至38.34 mm的任何位置。这些时间、频谱和空间特征可用于改进现有的伪迹抑制技术,激发新的伪迹减轻策略,并有助于开发新型皮层刺激方案。综上所述,这些发现加深了我们对皮层电刺激的理解,并为未来的BD-BCI系统提供了关键的设计规范。