Neural Recording and Simulation Tools to Address the Mesoscale Gap

摘要 我们设计了一种新的方法来跨区域和深度在大脑中执行多尺度记录。该工具因其方向性和可扩展感测而被称为DISC，是围绕电极导线主体的微电极阵列，旨在最大限度地提高“基底屏蔽”现象。电-准静态建模和体内数据证明了与微丝和环形电极相比的显著改进，包括（i）信号幅度，（ii）信噪比，和（iii）分类测试中的源分离。DISC以立体方式测量局部场电位，并具有显着的放大作用，这在中尺度隔离源方面特别强大。这项技术的传播的几个关键挑战是在MEMS制造方法的固有限制，我们仍然无法将局部场电位与详细的电路功能。为了解决这两个挑战，我们将开发一种基于气溶胶喷射打印的革命性微电子制造方法（目标1），并开发一种可以预测特定电压输出的生物物理模型（目标2）。当与生物物理正演模型相结合时，DISC的放大和方向性将是提高局部场势效用的独特和有力的工具。将使用慢性大鼠和猕猴听觉实验对硬件和软件工具进行确认。DISC将在音频刺激期间演示听觉核心中的层流和全网络记录。我们将分析DISC记录区分最佳频率电路的能力，并将其与各种虚拟宏电极进行对比，包括目前在sEEG中使用的环形电极。我们相信，这项多学科的工作将最终产生3个关键工具，可供神经科学和临床社区使用：（1）立体定向引导深度阵列，能够进行慢性，低噪声宽带记录，擅长高分辨率中尺度信息;（2）详细的，多尺度前向模型（NetPyNE/Brainstorm pipeline），产生特定于包括DISC在内的几种设备类型的模拟电压读数;（3）高分辨率反演模式，将源定位扩展到中尺度电压输入。软件开发将是开源的。

Abstract We have designed a novel approach to perform multi-scale recordings in the brain across regions and depths. This tool, referred to as DISC for its directional and scalable sensing, is an array of microelectrodes surrounding the lead body and designed to maximize the phenomenon of “substrate shielding”. Electro-quasistatic modeling and in vivo data demonstrate significant improvements over microwires and ring electrodes, including (i) signal amplitude, (ii) signal-to-noise ratio, and (iii) source separation in classification testing. DISC measures local field potentials in stereo and with significant amplification, which is especially powerful in isolating sources at the mesoscale. Several critical challenges for the propagation of this technology are the inherent limitations in photolithographic manufacturing methods, and our continuing inability to relate the local field potential with detailed circuit function. To address these two challenges, we will develop a revolutionary manufacturing method for microelectronics based on aerosol jet printing (Aim 1) and develop a biophysical model that can predict specific voltage outputs (Aim 2). The amplification and directionality of DISC when combined with biophysical forward models will be a unique and power tool to improve the utility of the local field potential. The validation of the hardware and software tools will be performed using chronic rat and macaque auditory experiments. DISC will demonstrate both laminar and network-wide recordings in the auditory core during audio stimulation. We will analyze the ability of DISC recordings to discriminate the best frequency circuits and contrast this with a variety of virtual macroelectrodes, including the ring electrode currently used in sEEG. We believe this multidisciplinary work will culminate in 3 critical tools being made available to the neuroscience and clinical communities: (1) a stereotactically-guided depth array capable of chronic, low-noise wideband recordings that excel at high-resolution mesoscale information; (2) a detailed, multi-scale forward model (NetPyNE/Brainstorm pipeline) that produces simulated voltage readings specific to several device types including DISC; and (3) a high-resolution inverse model, which will extend source localization to mesoscale voltage inputs. The software development will be open-source.