Feasibility Study of Biological Neural Networks Formations on a VLSI Chip

小行星9318328 本项目的目的是研究在VLSI芯片上人工培养的生物神经元形成的神经网络在信息处理方面的功能和特异性。 为了实现这一目标，需要了解神经突中信号传输和处理的方法，以及了解由生物神经元形成的神经网络的操作的方法。 本研究的目的有三：1）在超大规模集成电路芯片上，通过真实的时间测量，研究三卷Helisoma trivolvis的功能特异性神经元的动态神经元连接。 2)验证了突触素分子作为神经突细胞骨架的相邻相关分子在信息处理中发挥重要作用的假设。 3)在对实验结果进行数学建模的基础上，提出了一种非线性动态类赫布学习算法。 拟议中的研究将发展一门新学科，称为合成神经科学，即利用电子技术来帮助定量了解大脑，将神经科学和电子芯片工程结合在一起。

9318328 Kim The objective of this project is to investigate the functionality and the specificity of neural networks formed from artificially cultured biological neurons on a VLSI chip, with respect to information processing. To achieve the objective, methods to understand signal transmission and processing in neurites, as well as the methods to understand the operation of neural networks formed from biological neurons. For this project, there are three goals: 1) investigate the dynamic neuronal connectivity of function-specific neurons of the snail, Helisoma trivolvis on a VLSI chip with real time measurements. 2) test the hypothesis that the synapsin molecules play an important role on the information processing as the neighboring associated molecules of cytoskeleton of neurites. 3) develop a nonlinear dynamic Hebbian-like learning algorithm based on the mathematical modeling of the experimental results. The proposed research will develop a new discipline called synthetic neuroscience, i.e. the use of electronics technology to assist the quantitative understanding of the brian, bringing together neuroscience and electronic chip engineering.