NCS-FO: Fully Wireless Flexible Electrical-Acoustic Implant for High-Resolution Neural Stimulation and Recording at Large Scale

通过监测和调节大脑活动来绘制复杂脑回路的动态地图，可以增强我们对大脑功能的理解，并为更好地治疗和预防不同的神经系统疾病提供了希望。与大脑连接也有可能增强我们的感知、运动和认知能力，以及恢复因受伤或疾病而丧失的感觉和运动功能。开发具有全脑分布式神经回路高分辨率记录和刺激能力的闭环神经接口仍然是神经科学研究的一个重大挑战。目前的无创神经调节技术仍然存在空间分辨率较差（100-1000毫米3）的问题，而具有更精细分辨率的植入式方法通过高侵入性实质植入，只能提供100-1000毫米的有限覆盖范围。这一综合研究和教育计划实现了大规模（整个大脑）的高空间分辨率（200微米）的微创超声神经调节（和神经记录）。这个项目将为一套全面的神经接口提供一个独特的构建块。它将为神经科学开辟新的机会，在空间分辨率和动物脑刺激的覆盖范围方面取得重大进展。它还具有转化为人类临床应用的潜力，例如神经和精神疾病的治疗以及脑机接口。该项目还包括一个综合的外展和教育组成部分，以影响K-12教师和学生（特别是来自代表性不足的群体）、少数民族、本科生和研究生，并培养一支跨学科的劳动力队伍。这个项目将教育广大读者（特别是妇女）了解研究组成部分的科学和应用，并通过系统的排除故障活动提高他们的研究技能。不同学科的研究生课程也将通过相关的多学科项目和客座讲座进行转型。本项目包括科学研究在柔性平台（放置在脑表面，无实质穿透）上进行植入式超声刺激，同时提供高空间分辨率（200 μ m）和广泛覆盖范围（覆盖整个大脑），同时显著降低侵入性。这个多学科项目汇集了电气和生物医学工程以及材料、计算机和神经科学方面的专业知识，具有变革性，因为它可能是唯一一种承诺在不同深度、200微米高分辨率、无实质性植入的情况下对分布式大脑区域进行大规模刺激的方法，为在大时间和空间尺度上理解神经和认知系统开辟了新的场所。这项技术的发展建立在研究人员在电路、无线电源、柔性技术、薄膜超声阵列、机器学习和神经接口方面的优势之上。该项目通过研究一种灵活的、图像引导的（带有机器学习模型的）、混合电声可植入系统，该系统具有用于超声波刺激（和电生理记录）的柔性薄片（在大脑表面）的形状，从而突破了超声神经调节的极限。将解决三个基础研究空白。1)对于大规模和高分辨率超声光束聚焦和转向，将探索在高频（例如5 MHz）下扩大超声元件和专用集成电路（ASIC）通道数量的最佳方法。为了降低复杂性，将利用柔性衬底上的薄膜晶体管形成一个大型二维超声阵列，该阵列具有可选择的一维阵列（例如，256个元件），仅由一个ASIC驱动。2)将优化柔性衬底上薄膜晶体管开关的可选薄膜超声阵列，以实现高效率和高压输出。3)考虑到设备的灵活性（超声元件的方向）和植入后的效果，将开发基于图像序列分析的成像和机器学习模型来引导超声聚焦光束。在台式和体内设置的系统级演示将建立这种灵活的植入式系统的可行性。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Dynamic mapping of complex brain circuits by monitoring and modulating brain activity can enhance our understanding of brain functions and provide the promise of better treatment and prevention of different neurological disorders. Interfacing with the brain also has the potential to enhance our perceptual, motor, and cognitive capabilities, as well as to restore sensory and motor functions lost through injuries or diseases. The development of closed-loop neural interfaces with high-resolution recording and stimulation capabilities from the distributed neural circuits within the entire brain is still a grand challenge of neuroscience research. Current noninvasive neuromodulation techniques still suffer from poor spatial resolution ( 100-1000’s of mm3), while implantable methods with finer resolution only provide a limited coverage of 100-1000’s of neurons through highly invasive parenchymal implantation. This integrated research and education program enables minimally invasive ultrasound neuromodulation (and neural recording) of the brain with high spatial resolution ( 200 µm) at large scale (over the whole brain). This project will yield a unique building block for a comprehensive set of neural interfaces. It will open new opportunities in neuroscience with significant improvements in spatial resolution and coverage of the brain stimulation in animals. It will also have translational potential for clinical applications in humans, such as the treatment of neurological and psychiatric disorders and brain-machine interfaces. This project also includes an integrated outreach and educational component to impact K-12 teachers and students (particularly from underrepresented groups), minorities, and undergraduate and graduate students, and to develop an interdisciplinary workforce. This project will educate a broad audience (particularly women) in the science and applications of the research components and enhance their research skills through systematic troubleshooting activities. Graduate curriculums across different disciplines will also be transformed with related multidisciplinary projects and guest lectures.This project includes scientific research to investigate implantable ultrasound stimulation on a flexible platform (placed on the brain surface with no parenchymal penetration) to simultaneously provide high spatial resolution ( 200 µm) and broad coverage (over the whole brain) while dramatically reducing invasiveness. This multidisciplinary project, which brings together expertise in electrical and biomedical engineering as well as material, computer, and neuro science, is transformative in that it is potentially the only method that promises large-scale stimulation across distributed brain regions at different depths with high resolutions of 200 µm without parenchymal implantation, opening a new venue for understanding neural and cognitive systems at large temporal and spatial scales. The development of this technology builds upon investigators’ strength in circuits, wireless power, flexible technologies, thin-film ultrasound arrays, machine learning, and neural interfaces. The project pushes the limits of ultrasound neuromodulation by investigating a flexible, image-guided (with machine learning models), hybrid electrical-acoustic implantable system with the form factor of a thin flexible sheet (on the brain surface) for ultrasound stimulation (and electrophysiology recording). Three fundamental research gaps will be addressed. 1) For large-scale and high-resolution ultrasound beam focusing and steering, the optimal approach in scaling up the number of ultrasound elements and application-specific integrated circuit (ASIC) channels at high frequencies (e.g., 5 MHz) will be explored. To reduce the complexity, thin-film transistors on a flexible substrate will be leveraged to form a large two-dimensional ultrasound array with selectable one-dimensional arrays (e.g., 256-element) driven by only one ASIC. 2) Selectable thin-film ultrasound arrays with thin-film transistor switches on flexible substrate will be optimized to achieve high efficiency and high pressure output. 3) Imaging and machine learning models based on image sequence analysis will be developed to guide the ultrasound focused beam, considering the device flexibility (ultrasound elements’ orientation) and post-implantation effects. A system-level demonstration in benchtop and in vivo settings will establish the feasibility of this flexible implantable system.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.