Computational solution of spike overlapping using data-based subtraction algorithms to resolve synchronous sympathetic nerve discharge.
Computational solution of spike overlapping using data-based subtraction algorithms to resolve synchronous sympathetic nerve discharge.
复制标题
DOI：
10.3389/fncom.2013.00149
复制
发表时间：
2013
影响因子：
3.2
通讯作者：
Shyu LY
中科院分区：
文献类型：
--
作者：
Su CK;Chiang CH;Lee CM;Fan YP;Ho CM;Shyu LY
传达调节内脏功能的中枢命令的交感神经通常表现出同步爆发的活动。为了了解单个纤维如何同步放电，我们建立了“寡纤维记录技术”，以新生大鼠的体外内脏交感神经-胸脊髓制剂作为实验模型，同时记录“多个”神经纤维的活动。虽然可以很容易地从胶原酶解离的交感纤维记录到不同的尖峰电位，但同步神经放电引起的问题是尖峰重叠导致复杂波形的发生率更高。由于商业软件没有为尖峰重叠提供明确的解决方案，因此编写了一系列与 MATLAB 脚本结合的定制 LabVIEW 程序用于尖峰排序。波形特征提取后，尖峰被表示为数据点，并通过 k 均值聚类自动分组，然后进行主成分分析 (PCA)，以验证其波形均匀性。对于距簇质心超过霍特林 T2 距离的不同波形，使用独特的基于数据的减法算法 (SA) 来确定它们是否是将靠近簇质心的尖峰模式与原始记录中可观察到的其他信号叠加而产生的复杂波形。与商业软件相比，通过使用我们的算法对包含同步尖峰和复杂波形的合成数据进行分析，获得了更高的精度。此外，T2 选择的尖峰和 SA 检索的尖峰都合并为单位活动。进行定量分析以评估单位活动是否真正源自单纤维。我们的结论是，我们的程序的应用可以帮助解决同步交感神经放电（SND）。
Sympathetic nerves conveying central commands to regulate visceral functions often display activities in synchronous bursts. To understand how individual fibers fire synchronously, we establish “oligofiber recording techniques” to record “several” nerve fiber activities simultaneously, using in vitro splanchnic sympathetic nerve–thoracic spinal cord preparations of neonatal rats as experimental models. While distinct spike potentials were easily recorded from collagenase-dissociated sympathetic fibers, a problem arising from synchronous nerve discharges is a higher incidence of complex waveforms resulted from spike overlapping. Because commercial softwares do not provide an explicit solution for spike overlapping, a series of custom-made LabVIEW programs incorporated with MATLAB scripts was therefore written for spike sorting. Spikes were represented as data points after waveform feature extraction and automatically grouped by k-means clustering followed by principal component analysis (PCA) to verify their waveform homogeneity. For dissimilar waveforms with exceeding Hotelling's T2 distances from the cluster centroids, a unique data-based subtraction algorithm (SA) was used to determine if they were the complex waveforms resulted from superimposing a spike pattern close to the cluster centroid with the other signals that could be observed in original recordings. In comparisons with commercial software, higher accuracy was achieved by analyses using our algorithms for the synthetic data that contained synchronous spiking and complex waveforms. Moreover, both T2-selected and SA-retrieved spikes were combined as unit activities. Quantitative analyses were performed to evaluate if unit activities truly originated from single fibers. We conclude that applications of our programs can help to resolve synchronous sympathetic nerve discharges (SND).
登录
查看更多内容
影响因子：
14.8
作者：
Carlson, Greg C.;Coulter, Douglas A.
通讯作者：
Coulter, Douglas A.
影响因子：
2.5
作者：
Fee, MS;Mitra, PP;Kleinfeld, D
通讯作者：
Kleinfeld, D
影响因子：
2.5
作者：
Mackenzie, PJ;Murphy, TH
通讯作者：
Murphy, TH
影响因子：
2.5
作者：
Häbler, HJ;Bartsch, T;Jänig, W
通讯作者：
Jänig, W
影响因子：
5.5
作者：
Miles, GB;Dai, Y;Brownstone, RM
通讯作者：
Brownstone, RM