EAGER: Towards High-throughput Nanophotonic Brain-Machine Interfaces

1555720（Jornet）几十年来，人类和机器之间的交互一直局限于视觉，听觉和触觉信息的交换。对现有人机接口（HMI）的概念分析表明，它们可以传输的有用信息量通常不受人脑或机器处理器的能力限制，而是受它们之间的接口限制，例如处理视觉，听觉和触觉信息所需的传感器器官。对于患有发育和衰老相关残疾的人来说尤其如此，他们的感觉器官或肌肉骨骼系统进一步限制了传统HMI的功能。为了克服这种限制，在过去的十年中，已经提出了几种脑机接口（BMI），它们在大脑和远程机器之间建立了直接的路径。例如，脑电图（EEG）信号已经成功地用于以非侵入性方式并且以高时间分辨率控制机器。然而，基于EEG的BMI不能用于读取单个神经元的活动，而只能读取它们的集体反应。类似地，基于光遗传学的BMI依赖于使用光与大脑中的遗传修饰神经元相互作用，可以用于更准确地读取或控制大脑中的神经元活动。然而，现有的用于BMIs的光学器件具有高度侵入性，并且难以与单个神经元连接。在本项目中，将利用纳米光子学，纳米电子学和无线通信的最新技术开发新型纳米光子BMIs。所提出的技术依赖于使用纳米设备的分布式网络来以非常高的空间和时间精度以及微创方式监测和控制人脑的神经元活动。这项技术有望显著改变人类与机器互动的方式，并通过为残疾人提供与环境互动的变革性方式，恢复功能能力和认知能力，显著改善残疾人的生活质量。该项目的目的是证明新型纳米光子脑机接口的可行性，该接口基于使用纳米设备的分布式网络来监测和控制大脑中的神经元活动。其基本思想是用一个协调的纳米器件网络取代现有的微发光二极管阵列和微光电探测器阵列，这些纳米器件能够光学地激发单个神经元并测量它们的活动。这种方法的好处有几个。首先，光学纳米天线的尺寸非常小，最大尺寸小于1微米，使得能够以非常高的精度测量单个神经元中的神经元活动。此外，每个单独的纳米器件的总尺寸小于几十立方微米，从而最大限度地减少了这种方法的侵入性。此外，通过在光学频率下操作，可以实现非常高的时间分辨率，这可以实现对神经元活动中的高频时间瞬变的测量。在这个长期目标中，这个为期两年的EAGER项目的重点是为下一代纳米光子脑机接口建立分布式神经元活动监测的基础。该项目将沿着以下三个主要方向：i）设计光学纳米天线，用于有效检测神经元活动产生的可见电磁辐射; ii）开发实验光遗传学的神经元平台; iii）纳米光子BMI的系统级设计指南。就更广泛的影响而言，该项目有望为高通量纳米光子BMI的开发铺平道路。所提出的方法可以显着简化和降低现有的单神经元监测和控制平台的成本，提高空间和时间分辨率。纳米光子BMI有可能显著改善残疾人的生活质量，为他们提供一种与机器双向互动的新方式，并最终与他们的环境互动。特别是，在大脑和外部机器之间建立一条“直接通道”，可以帮助克服一般或与年龄有关的残疾人的局限性，恢复人类的功能能力，甚至认知能力。例如，来自大脑的神经信号可以用来直接控制计算机甚至外骨骼。同样，所提出的技术可以帮助开发许多发育和衰老相关疾病的变革性治疗方法，例如阿尔茨海默病或精神分裂症，其起源在于连续神经元之间的通信问题。

1555720(Jornet)For many decades, the interaction between humans and machines has been restricted to the exchange of visual, auditory and tactile information. A conceptual analysis of the existing human-machine interfaces (HMIs) reveals that the amount of useful information that they can transfer is generally not limited by the capabilities of the human brain or those of the machine processor, but by the interfaces between them, such as the sensor organs that are required to handle visual, auditory and tactile information. This is especially true for people with developmental- and aging-related disabilities, whose sensor organs or musculoskeletal system further limit the functionality of traditional HMIs. To overcome such limitation, several brain-machine interfaces (BMIs), which establish a direct path between the brain and a remote machine, have been proposed in the last decade. For example, electroencephalogram (EEG) signals have been successfully utilized to control machines in a non-invasive way and with high temporal resolution. However, EEG-based BMIs cannot be utilized to read the activity from individual neurons, but only their collective response. Similarly, optogenetics-based BMIs, which rely on the use of light to interact with genetically modified neurons in the brain, can be utilized to more accurately read or control the neuronal activity in the brain. However, existing optical devices used for BMIs are highly invasive and difficult to interface with single neurons.In this project, novel nanophotonic BMIs will be developed by leveraging the state of the art in nanophotonics, nanoelectronics and wireless communications. The proposed technology relies on the use of a distributed network of nano-devices to monitor and control the neuronal activity of the human brain with very high spatial and temporal accuracy and in a minimally invasive way. This technology is expected to significantly change the way in which humans interact with machines and can significantly improve the quality of life of people with disabilities, by providing them a transformative way to interact with the environment and restoring functional abilities as well as cognition capabilities. The objective of the proposed project is to prove the feasibility of novel nanophotonic brain-machine interfaces based on the use of a distributed network of nano-devices to monitor and control the neuronal activity in the brain. The fundamental idea is to replace existing micro-led arrays and micro-photodetector arrays by a network of coordinated nano-devices, which are able to optically excite individual neurons and measure their activity. The benefits of this approach are several. First, the very small size of optical nano-antennas, below one micrometer in the largest dimension, enables the possibility to measure the neuronal activity in a single neuron with very high accuracy. In addition, the total size of each individual nano-device is expectedly below several tens of cubic micrometers, thus minimizing the invasiveness of this approach. Moreover, by operating at optical frequencies, a very high temporal resolution is possible, which can enable the measurement of high-frequency time-transients in the neuronal activity.Within this long-term goal, the focus of this two-year EAGER project is on establishing the foundations of distributed neuronal activity monitoring with cooperative nano-devices for next-generation nanophotonic brain-machine interfaces. Contributions will be made along the following three main thrusts: i) Design of optical nano-antennas for efficient detection of visible electromagnetic radiation generated by neuronal activity; ii) Development of a neuronal platform for experimental optogenetics; and, iii) System-level design guidelines for nanophotonic BMIs.In terms of broader impact, the project is expected to pave the way for the development of high-throughput nanophotonic BMIs. The proposed approach can significantly simplify and reduce the cost of existing single-neuron monitoring and control platforms, with increased spatial and temporal resolutions. Nanophotonic BMIs have the potential to significantly improve the quality of life of people with disabilities, by providing them a new way to bidirectionally interact with machines and, ultimately, their environment. In particular, the creation of a "direct-path" between the brain and external machines can help to overcome the limitations of people with general or aging-related disabilities and restore human functional abilities and even cognition. For example, neural signals from the brain could be utilized to directly control a computer or even an exoskeleton. Similarly, the proposed technology could help to develop transformative treatments for many developmental- and aging-related diseases, such as Alzheimer's disease or Schizophrenia, whose origin lies at communication problems between consecutive neurons.