大鼠脑神经元锋电位诱发采集与分析软件系统研究

发布时间：2018-01-15 09:24

本文关键词：大鼠脑神经元锋电位诱发采集与分析软件系统研究　出处：《新乡医学院》2015年硕士论文　论文类型：学位论文

【摘要】：神经接口(Neural Interface, NI)是连接外界环境与神经系统的通道,外部世界的信息可以通过NI转换成特定模式的电刺激输入神经系统,用于改善、修复神经系统的某些缺陷,也可以通过NI将神经系统中的信息提取出来,经过信号处理,翻译成机器可以识别的命令来控制外部设备。生物机器人是体内植入了神经接口的动物,通过神经接口可以将外部控制命令转化为电刺激输入动物神经系统,控制动物的行为,同时也可以通过神经接口将动物神经系统中的神经元电信号提取出来,经过特征提取以及模式分类,可以观察到动物的生理状态。目前,应用于生物机器人的神经接口大多数只是单向信息传递。单输入型神经接口无法定量评估刺激对动物的控制作用,单输出型神经接口由于缺乏刺激,无法控制动物的行为。因此,为了解决神经激励-响应模型、刺激参数的定量评价以及激励参数的优化等问题,实现对生物机器的有效精确控制,研究小型且具有双向信息传递功能的神经接口技术具有越来越大的研究价值。本文设计并搭建了基于神经接口技术的生物机器人软件系统平台。本文首先对神经接口技术和生物机器人进行了详细的介绍,揭示了神经接口技术研究的重要性和生物机器人研究的紧迫性,并提出了具体的研究内容。然后介绍了实现生物机器人的技术基础,包括了电极制作、动物手术、Visual Studio 2010编程技术、Android智能手机开发平台以及神经元电信号处理技术。其次本文提出了基于双向神经接口的生物机器人系统框架,并详细介绍和实现了系统中的刺激参数设置模块、蓝牙设备连接模块、基于Android智能手机的刺激软件、神经元电信号解析模块、实时信号采集模块和神经元信号处理模块。刺激参数设置模块和蓝牙设备连接模块实现了输入型神经接口,可以将实验人员设置的命令翻译成对应模式的电刺激输入大鼠脑神经,控制大鼠行为：神经元电信号解析模块和神经元信号处理实现了输出型神经接口,可以对大鼠神经元发放的电信号进行采集以及分析处理。此外,本文设计实验测试了生物机器人系统的性能。首先通过设计一个测试软件验证了输入系统的连通性；其次,本文利用加噪后的模拟信号对算法性能进行测试；然后,本文利用改进后的算法对大鼠伸缩实验范式下的脑神经信号进行分类,得到了满意的结果。最后,对全文进行总结,并指出在今后的科研工作中需要继续努力研究的问题。
[Abstract]:Neural interface (NI) is a channel that connects the external environment with the nervous system. The information of the outside world can be transformed into a specific mode of electrical stimulation into the nervous system through NI, which can be used to improve and repair some defects of the nervous system, but also through NI to extract the information from the nervous system. After signal processing, it is translated into a machine-recognizable command to control an external device. A biological robot is an animal with a neural interface implanted in the body. Through the neural interface, the external control command can be transformed into electrical stimulation into the animal nervous system, and the behavior of the animal can be controlled. At the same time, the neural interface can also be used to extract the neuron electrical signals from the animal nervous system. After feature extraction and pattern classification, the physiological state of animals can be observed. Most of the neural interfaces used in biological robots are only one-way information transmission. Single input neural interface can not quantitatively evaluate the control effect of stimulation on animals, and single output neural interface is lack of stimulation. Therefore, in order to solve the problems of neural excitation-response model, quantitative evaluation of stimulation parameters and optimization of stimulation parameters, the effective and accurate control of biological machinery can be realized. It is more and more valuable to study the neural interface technology which has the function of bidirectional information transmission. This paper designs and builds the software system platform of the biological robot based on the neural interface technology. Interface technology and biological robot are introduced in detail. This paper reveals the importance of neural interface technology and the urgency of biological robot research, and puts forward the specific research contents. Then, it introduces the technical basis of realizing biological robot, including electrode fabrication. Visual Studio 2010 programming technique for animal surgery. Android smart phone development platform and neuron signal processing technology. Secondly, this paper proposes a bidirectional neural interface based biological robot system framework. And detailed introduction and implementation of the system stimulation parameter setting module, Bluetooth device connection module, based on Android smart phone stimulation software, neuron signal analysis module. The real-time signal acquisition module and neuron signal processing module, the stimulation parameter setting module and the Bluetooth device connection module realize the input neural interface. The commands set by the experimenter can be translated into electrical stimulation of corresponding mode into the rat brain nerve to control the behavior of the rats: the neuronal electrical signal analysis module and the neuronal signal processing implement the output neural interface. The electrical signals issued by rat neurons can be collected and analyzed. In this paper, experiments are designed to test the performance of the biological robot system. Firstly, a test software is designed to verify the connectivity of the input system. Secondly, the performance of the algorithm is tested by using the noised analog signal. Then, the improved algorithm is used to classify the neural signals in the rat extensional experiment paradigm, and the results are satisfactory. Finally, the paper summarizes the full text. And pointed out that in the future scientific research work need to continue to study the problems.
【学位授予单位】：新乡医学院
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP242;R49

【参考文献】