Lopour Beth A, Staba Richard J, Stern John M, Fried Itzhak, Ringach Dario L
Department of Biomedical Engineering, University of California, Irvine, CA 92697, USA.
J Neural Eng. 2016 Apr;13(2):026031. doi: 10.1088/1741-2560/13/2/026031. Epub 2016 Mar 15.
Quantifying the relationship between microelectrode-recorded multi-unit activity (MUA) and local field potentials (LFPs) in distinct brain regions can provide detailed information on the extent of functional connectivity in spatially widespread networks. These methods are common in studies of cognition using non-human animal models, but are rare in humans. Here we applied a neuronal spike-triggered impulse response to electrophysiological recordings from the human epileptic brain for the first time, and we evaluate functional connectivity in relation to brain areas supporting the generation of seizures.
Broadband interictal electrophysiological data were recorded from microwires adapted to clinical depth electrodes that were implanted bilaterally using stereotactic techniques in six presurgical patients with medically refractory epilepsy. MUA and LFPs were isolated in each microwire, and we calculated the impulse response between the MUA on one microwire and the LFPs on a second microwire for all possible MUA/LFP pairs. Results were compared to clinical seizure localization, including sites of seizure onset and interictal epileptiform discharges.
We detected significant interictal long-range functional connections in each subject, in some cases across hemispheres. Results were consistent between two independent datasets, and the timing and location of significant impulse responses reflected anatomical connectivity. However, within individual subjects, the spatial distribution of impulse responses was unique. In two subjects with clear seizure localization and successful surgery, the epileptogenic zone was associated with significant impulse responses.
The results suggest that the spike-triggered impulse response can provide valuable information about the neuronal networks that contribute to seizures using only interictal data. This technique will enable testing of specific hypotheses regarding functional connectivity in epilepsy and the relationship between functional properties and imaging findings. Beyond epilepsy, we expect that the impulse response could be more broadly applied as a measure of long-range functional connectivity in studies of cognition.
量化不同脑区中微电极记录的多单元活动（MUA）与局部场电位（LFP）之间的关系，可为空间广泛网络中功能连接程度提供详细信息。这些方法在使用非人类动物模型的认知研究中很常见，但在人类研究中很少见。在此，我们首次将神经元尖峰触发脉冲响应应用于人类癫痫脑的电生理记录，并评估与支持癫痫发作产生的脑区相关的功能连接。
从适配于临床深度电极的微丝记录宽带发作间期电生理数据，这些电极通过立体定向技术双侧植入6例药物难治性癫痫的术前患者体内。在每根微丝中分离出MUA和LFP，并针对所有可能的MUA/LFP对，计算一根微丝上的MUA与另一根微丝上的LFP之间的脉冲响应。将结果与临床癫痫定位进行比较，包括癫痫发作起始部位和发作间期癫痫样放电部位。
我们在每个受试者中检测到显著的发作间期远程功能连接，在某些情况下跨半球。两个独立数据集的结果一致，显著脉冲响应的时间和位置反映了解剖学连接。然而，在个体受试者中，脉冲响应的空间分布是独特的。在两名癫痫定位明确且手术成功的受试者中，致痫区与显著的脉冲响应相关。
结果表明，尖峰触发脉冲响应仅使用发作间期数据就能提供有关促成癫痫发作的神经元网络的有价值信息。该技术将能够测试关于癫痫中功能连接以及功能特性与影像学发现之间关系的特定假设。除癫痫外，我们预计脉冲响应可更广泛地应用于认知研究中作为远程功能连接的一种测量方法。
