Mino Hiroyuki, Rubinstein Jay T
Department of Electrical and Computer Engineering, Kanto Gakuin University, Yokohama 236-8501, Japan.
IEEE Trans Neural Syst Rehabil Eng. 2006 Sep;14(3):273-80. doi: 10.1109/TNSRE.2006.881590.
In this paper, the effects of neural refractoriness on action potential (spike) initiations with electrical stimulation are investigated using computer modeling and simulation techniques. The computational model was composed of a myelinated nerve fiber with 50 nodes of Ranvier, each consisting of stochastic sodium and potassium channels, making it possible to represent the fluctuations of spike initiation. A series of two-pulse stimuli was presented by a stimulating electrode above the central (26th) node of Ranvier. The amplitude of the first (masker) pulse stimulus was set such that the masker pulse stimulus evoked spikes on each trial, while that of the second (probe) pulse stimulus was set such that the probe pulse stimulus evoked spikes on a half of trials, threshold values. Then the transmembrane potentials in response to the probe pulse stimulus were recorded at each node (i.e., 1-50 nodes) in order to determine the spike initiation node and time. From the observation of the spike initiation node and time, a spatio-temporal histogram as well as a spatial variability and a temporal variability of spike initiations was generated which allowed us to interpret fluctuations in spike initiation node and time. It was shown that the distribution of spike initiations tended to become greater spatially and longer temporally as the masker-probe intervals (MPIs) of the two-pulse stimuli shortened. It was also shown that the number of activated sodium channels as functions of space and time tended to become smaller due to inactivation of sodium channels and varied spatially and temporally as MPIs shortened. These findings may imply that the stochastic sodium channels during a relative refractory period may contribute to enhancing the fluctuations in spike initiations, and give us an insight into encoding information with electric stimuli to improve the performance of the prosthetic devices, especially cochlear implants.
在本文中，利用计算机建模和仿真技术研究了神经不应期对电刺激引发动作电位（尖峰）的影响。计算模型由一条具有50个郎飞结的有髓神经纤维组成，每个郎飞结都包含随机的钠通道和钾通道，从而能够呈现尖峰起始的波动情况。在郎飞结中央（第26个）节点上方通过刺激电极施加一系列双脉冲刺激。设置第一个（掩蔽）脉冲刺激的幅度，使得掩蔽脉冲刺激在每次试验中都能引发尖峰，而设置第二个（探测）脉冲刺激的幅度，使得探测脉冲刺激在一半的试验中能引发尖峰，即阈值。然后在每个节点（即1 - 50个节点）记录对探测脉冲刺激的跨膜电位，以确定尖峰起始节点和时间。通过观察尖峰起始节点和时间，生成了时空直方图以及尖峰起始的空间变异性和时间变异性，这使我们能够解释尖峰起始节点和时间的波动情况。结果表明，随着双脉冲刺激的掩蔽 - 探测间隔（MPI）缩短，尖峰起始的分布在空间上趋于更分散，在时间上趋于更长。还表明，由于钠通道失活，作为空间和时间函数的激活钠通道数量趋于减少，并且随着MPI缩短在空间和时间上发生变化。这些发现可能意味着相对不应期的随机钠通道可能有助于增强尖峰起始的波动，并为我们深入了解利用电刺激编码信息以提高假体装置，特别是人工耳蜗的性能提供了思路。
