The Detection And Analysis Of Neural Activity Of Visual Cortex Through Duramater

Vision system is the most complicated structure and fuction in human body and visual performance is very important for human being to obtain outside information, and it deeply influences human life, study and work. However, there are 40-50 million people who have been suffering visual handicapped and the number is also increasing.Therefore it has both scientific and practical significance on studying vision rehabilitation. In order to develop visual cortex prosthesis based on dural mate and perform effectively stimulating visual cortex, the first step is to identify what is the expression on the dura during visual cortical neural excition evoked the visual stimulation, which can be used to instruct to encode signal of electrical stimulation ouside visual cortex.In this paper, the animal experimental platform was set up basing on the flexible microelectrode array and using multi-channel physiological signals recording system, which has been used to record the neural activity of visual cortex of cat and Macaca thibetana. The expeimental data was processed and analyzed in time-domain, frequency-domain and spatial-domain by several classic signal processing and anlysis methods. The signal processing and analysis methods in time-domain we used contain superimposed average technology, zero-phase filtering and extracting the amplitude and incubation period. In frequency domain, we maily used spectum analysis. The result showed that: (1) the neural activity signal of visual cortex had been similar with the typical VEP on waveform characteristics when the stimuli frequency is lower; (2) the amplitude spectrum gained by spectral analysis presented apparent amplitude peak at the corresponding frequencies of visual stimulations when the frequencies of visual stimulations are higher (16Hz, 20Hz, 24Hz); (3) there were differences between the VEP recorded in different visual cortex domain responding to microelectrode evoked by the same visual stimulations; the excited models of visual cortical are also different for visual stimulations from different visual fields; (4) the time and spatial resolution of visual stimulation effect the amplitude and latent period of P wave of epidural VEP.In addition, to carry out the electrostimulation experiment study on the epidural neural activity, 64-channel bipolar current pluse signal generator using as cortical electrical pluse generator was designed and developed by FPGA system and peripheral electrical circuit. The voltage pluse signals and the selection of ouput channels were finished by FPGA system, and ne-way voltage pluse was converted to bipolar pluse current signal by peripheral electrical circuit. Through the result of software emulation and hardware testing, it is found that this pluse development system achieved the pre-design requirements and its function of outputting 64-channel bipolar electrical pulse signal, whereas it occupies less FPGA resources and the number of stimulating channels increases greatly.The Visual Evoked Potential were recorded in this research by the flexible microelectrode array fixed on the dura, and the result that the visual cortex own different excited models responding to different visual stimulations from different visual fields was detected, and these results provide experimental basis for the development of visual cortex prostheses; We also verified that the recording of the electrical activity of the cerebral cortex on the dura was a neural electrophysiology technique with characters of minimal invasive, high reliability and high spatial and temporal resolution, and this provide a better technical means of detecting of the activity of visual cortex and other cortices. In addition, the 64-channel electrical stimulation signal generator designed in this research is powerful and outputs stably and has important practical significance for the development of artificial visual prostheses.