NCS-FO: A Computational Theory to Model the Neurobiological Basis of a Visuo-Cognitive Neuroprosthetic

通过对大脑的直接微刺激来唤起盲人高质量的视觉感受是非常困难的。一个主要的障碍是，大脑的电刺激通常会影响相互抑制的神经元群，从而破坏正常的神经元信号。视觉系统有两个对抗性的信息通道，它们编码对亮度的感知，在开细胞中，或在关细胞中。这两个通道不适当的共同激活导致对比度无效，并剥夺视觉感知。因此，高质量的假肢刺激系统必须避免不必要的相互抑制神经元的协同激活，就像自然视觉系统一样。这是一个挑战，因为在大脑中，拮抗神经元通常位于彼此微米范围内。该项目旨在在病毒转染、成像方法、计算理论和皮质假体神经工程设计方面取得革命性进展，目的是通过对大脑中的神经元进行基因修饰，然后用光刺激它们来恢复视力，这种方法被称为光遗传学。预期的结果和方法将构成构建突破性神经假肢的科学基础，具有进一步研究大脑皮层的感觉、运动和认知部分的变革性潜力，并推动人类医学发展。为了促进衍生产品的发展和向公众提供，研究小组将通过公开讲座和在科普杂志上发表出版物向公众传播这些发现。研究人员将监督来自代表性不足群体的受训者，包括博士后研究员、研究生、本科生和高中生。调查人员是儿童援助协会（CAS）劳动力发展部暑期青年就业计划（SYEP）的教师导师，该计划为纽约市的弱势和少数族裔青年提供暑期研究机会，以激励他们追求STEM职业。最近的研究表明，对于任何给定的视网膜位置，大脑视觉皮层的“开”和“关”细胞输入纯粹是兴奋的，集中在特定的一层，并以一种可以被大脑外部光线瞄准的模式排列。首先，研究小组将对这些神经元进行基因改造，把它们变成一种新型的光感受器，嵌入大脑。然后，该团队将对光刺激识别的ON和OFF细胞，确定激活任一通道的精确平衡，从而根据摄像机的信号生成高质量的假肢视觉。然后，这项技术可以用来绕过眼睛，通过摄像头刺激大脑。该项目旨在开发计算模型来驱动光遗传脑刺激系统，以最佳方式激活初级视觉皮层的神经反应。通过比较视力正常的非人类灵长类动物对自然视觉刺激的神经元反应与同一神经元中假肢的反应，这项工作将优化激发自然视觉感知的刺激模式。在视觉空间的每个位置上，适当的ON和OFF输入的平衡目标有望在最高可达到的敏锐度下实现最大的对比度感知，具有完整的立体双目视觉。该团队的时空视觉输入到大脑皮层的计算模型也将解释眼球运动对早期视觉反应的影响，这是一种首次在这里测试的视觉假肢的新方法。该项目由理解神经和认知系统的综合策略（NSF-NCS）资助，这是一个由计算机与信息科学与工程（CISE）、教育与人力资源（EHR）、工程（ENG）和社会、行为和经济科学（SBE）联合支持的多学科项目。

Evoking high quality visual perception in a blind person, via direct microstimulation of the brain, poses great difficulties. One major obstacle has been that electrical stimulation of the brain typically affects neuronal populations that are mutually suppressive, which subverts proper neuronal signaling. The visual system has two antagonistic information channels that encode either the perception of lightness, in ON cells, or darkness, in OFF cells. Inappropriate coactivation of these two channels results in nullification of contrast, and deprived visual perception. It follows that high-quality prosthetic stimulation systems must avoid unwanted coactivation of mutually suppressive neurons, just as the natural visual system does. This is a challenge because the antagonistic neurons typically lie within microns of each other in the brain. The project aims at transformative advances in viral transfection and imaging methodology, computational theory, and cortical prosthetic neuroengineering design for the purpose of restoring vision by genetically modifying neurons in the brain and then stimulating them with light, a method called optogenetics. The expected results and methodology will form the scientific basis to build a breakthrough neuroprosthetic, with transformative potential to further brain research in sensory, motor, and cognitive parts of the cortex and to advance human medicine. To promote the development and availability of derived products to the public, the team will disseminate the discoveries to general audiences through public lectures and publications in popular science magazines. The investigators will supervise trainees from underrepresented groups, including postdoctoral fellows, graduate students, undergraduates, and high school students. The investigators are faculty mentors for The Children's Aid Society (CAS) Workforce Development Department Summer Youth Employment Program (SYEP), which provides summer research opportunities to disadvantaged and minority youth in NYC to inspire them to pursue STEM careers. Recent research has shown that, for any given retinal position, the ON and OFF cell inputs to the brain's visual cortex are purely excitatory, concentrate in a specific layer, and are laid out in a pattern that can be targeted with light from outside the brain. First, the team will modify these neurons genetically, to turn them into a novel type of photoreceptor, embedded within the brain. The team will then target light stimulation to the identified ON and OFF cells, determining the precise balance of activation to either channel to generate high-quality prosthetic vision based on a video camera's signal. This technology can then be used to bypass the eye to stimulate the brain from the camera. The project aims to develop the computational model to drive an optogenetic brain stimulation system that will optimally activate neural responses in the primary visual cortex. By comparing the neuronal responses of sighted nonhuman primates viewing natural visual stimuli to prosthetic responses in the same neurons, the work will optimize stimulation patterns that evoke naturalistic visual perception. The balanced targeting of appropriate ON and OFF inputs at each position in visual space is expected to achieve maximal contrast perception at the highest attainable acuity, with full stereoscopic binocular vision. The team's computational model of spatiotemporal visual inputs into the cortex will also account for the effects of eye movements on early visual responses, a novel approach to visual prosthetics tested here for the first time.This project is funded by Integrative Strategies for Understanding Neural and Cognitive Systems (NSF-NCS), a multidisciplinary program jointly supported by the Directorates for Computer and Information Science and Engineering (CISE), Education and Human Resources (EHR), Engineering (ENG), and Social, Behavioral, and Economic Sciences (SBE).