视网膜假体视觉信息处理模型

Modeling of Visual Information Processing in Retinal Prosthesis

视觉假体方案是神经工程领域用于部分或完全恢复视觉的主要手段,建立生物视网膜模型,并仿真其视觉信息处理功能,是视网膜假体研究中的一个重要组成部分。从已知的生理机制出发,提出两层结构的视网膜信息处理模型:外网状层(Out Plexiform Layer,OPL)信息提取和内网状层(Inner Plexiform Layer,IPL)信息编码,通过时空滤波、静态非线性调整和泊松峰电位产生,建立输入刺激图像和输出峰电位序列之间的直接关系,并在MATLAB平台上结合其图形模块进行仿真研究,得到了与内、外网状层处理结果对应的边缘轮廓图像和携带视觉信息的峰电位序列,为视网膜假体研究提供了一种理论上的可行性模型。

Visual prosthesis offers alternative way for partially repairing visual impairment.The Modeling and simulating of how biological retina encodes visual information with spike trains is a key issue for the development of a retinal prosthesis.Based on state-of-the-art retinal physiological mechanism,a two-layered visual information model is brought forward.It includes the Out Plexiform Layer（OPL） for information extraction and the Inner Plexiform Layer（IPL） for information encoding.Using spatiotemporal filter,static nonlinear rectification and Poisson spike generation,we establish the relations between input stimulus images and output spike trains.The simulation result shows us the contour edge images and spike sequences corresponding stimulus images.A theory model to develop artificial retina for the retinally blind is given.