Li Xiaoli, Li Duan, Voss Logan J, Sleigh Jamie W
Institute of Electrical Engineering, Yanshan University, Qinhuangdao, Hebei, 066004, China.
Neuroimage. 2009 Nov 15;48(3):501-14. doi: 10.1016/j.neuroimage.2009.07.008. Epub 2009 Jul 14.
Brain functions are related to neuronal networks of different sizes and distribution, and neuronal networks of different sizes oscillate at different frequencies. Thus the synchronization of neuronal networks is often reflected by cross-frequency interaction. The description of this cross-frequency interaction is therefore a crucial issue in understanding the modulation mechanisms between neuronal populations. A number of different kinds of interaction between frequencies have been reported. In this paper, we develop a general harmonic wavelet transform based bicoherence using a phase randomization method. This allows us to measure the comodulation of oscillations between different frequency bands in neuronal populations. The performance of the method is evaluated by a simulation study. The results show that the improved wavelet bicoherence method can detect a reliable phase coupling value, and also identify zero bicoherence for waves that are not phase-coupled. Spurious bicoherences can be effectively eliminated through the phase randomization method. Finally, this method is applied to electrocorticogram data recorded from rats during transitions between slow-wave sleep, rapid-eye movement sleep and waking. The phase coupling in rapid-eye movement sleep is statistically lower than that during slow-wave sleep, and slightly less than those in the wakeful state. The degree of phase coupling in rapid-eye movement sleep after slow-wave sleep is greater than in rapid-eye movement sleep prior to waking. This method could be applied to investigate the cross-frequency interactions in other physiological signals.
脑功能与不同大小和分布的神经元网络相关，且不同大小的神经元网络以不同频率振荡。因此，神经元网络的同步性常通过交叉频率相互作用来体现。所以，对这种交叉频率相互作用的描述是理解神经元群体间调制机制的关键问题。已有多种不同频率间相互作用的报道。在本文中，我们使用相位随机化方法开发了一种基于广义谐波小波变换的双相干性方法。这使我们能够测量神经元群体中不同频段振荡之间的共调制。通过模拟研究对该方法的性能进行了评估。结果表明，改进后的小波双相干性方法能够检测到可靠的相位耦合值，还能识别非相位耦合波的零双相干性。通过相位随机化方法可有效消除虚假双相干性。最后，将该方法应用于大鼠在慢波睡眠、快速眼动睡眠和清醒状态转换期间记录的皮质脑电图数据。快速眼动睡眠中的相位耦合在统计学上低于慢波睡眠期间，且略低于清醒状态下的相位耦合。慢波睡眠后快速眼动睡眠中的相位耦合程度大于清醒前快速眼动睡眠中的相位耦合程度。该方法可用于研究其他生理信号中的交叉频率相互作用。
