视网膜神经节细胞非经典感受野的模型研究与脑信号的复杂度分析

【作者】 曹洋；

【导师】 顾凡及；

【作者基本信息】 复旦大学 ， 生物物理学， 2005， 博士

【摘要】 建立了视网膜神经节细胞的三层网络模型，以检验李朝义等提出的非经典感受野可能源于无长突细胞相互抑制的假说，并探讨了神经节细胞感受野的方位倾向性的可能机制。模拟结果表明，通过无长突细胞之间的相互抑制可以形成神经节细胞的非经典感受野。模型模拟了神经节细胞的中心区和大周边区的方位倾向性及其相互作用，结果提示神经节细胞的方位倾向性可能主要源于神经节细胞的椭圆形感受野范围。利用本实验室自己提出的二维C₀复杂度，对内源性光学成像技术得到的猫初级视皮层的方位功能图进行了详细地分析。结果表明，二维C₀复杂度能够准确地反映内源性光学成像方位功能图出现和消退的过程，能够定量描述急性升高眼内压对初级视皮层方位功能图的影响，能够定量描述可逆性失活皮层后内侧外上薛氏回区对初级视皮层方位功能图的影响。结果提示，二维C₀复杂度可以作为客观描述皮层功能状态的定量指标。利用本实验室自己提出的二阶近似熵复杂度，对处于昏迷状态和脑死亡状态的病人的脑电图进行了初步的分析和比较。结果表明，昏迷和脑死亡脑电图的二阶近似熵复杂度存在着显著的差异，这表明脑死亡脑电图的非平稳性显著的降低了。结果提示，脑电图非平稳性的显著降低可能是脑死亡脑电图的特征之一。更多还原

【Abstract】 To test the assumption that the non-classical receptive field of retinal ganglion cells might be formed from the mutual inhibition among the amacrine cells, and to explore the possible mechanism underlying the orientation bias, a three-layer model of retinal ganglion cells was constructed. Both area responses and orientation bias were simulated. Results hint that mutual inhibition among amacrine cells may play a key role in the formation of extended surround in the non-classical receptive field of the retinal ganglion cell, and the orientation bias may mainly be decided by the elliptic shape of thereceptive field of the ganglion cell.2D Co complexity was used to analyze the orientation maps recorded by optical imaging based on intrinsic signals, which is a powerful method to view the activities of neural assembly in the cortex of animals in vivo, especially the detailed functional architecture of visual cortex. Results have shown that 2D Co complexity could reveal the occurrence of the orientation preference map in the visual cortex, and describe the variance of the neural responses in the primary visual cortex under high intraocular pressure, and under the reversible inactivation of PLMS. This suggests 2D Co complexity could be used as a new quantitative measure for the intrinsic optical images.The 2^nd order ApEn complexity of EEG signals from a patient in coma and in brain death was estimated. Results show that compared with coma state, the 2^nd order ApEn complexity under brain death state is significant reduced, which suggested the significant reduction of nonstationarity degree may be one of the characteristics of EEG during brain death.更多还原