Bhattacharya Joydeep, Pereda Ernesto
Department of Psychology, Goldsmiths College, University of London, New Cross, SE14 6NW, London, UK.
Commission for Scientific Visualization, Austrian Academy of Sciences, Donau City Str. 1, A1220, Vienna, Austria.
J Comput Neurosci. 2010 Aug;29(1-2):13-22. doi: 10.1007/s10827-009-0155-5. Epub 2009 May 6.
Irregular and complex signals are ubiquitous in nature. The principal aim of this paper is to develop an index, quantifying the complexity of such signals, which is based on the distribution of the strengths of its orthogonal oscillatory modes estimated by singular value decomposition. The index is first tested with simulated chaotic and/or stochastic maps and flows. Among neural data analysis, the index is first applied to a cognitive EEG data set recorded from two groups, musicians and non-musicians, during listening to music and resting state. In the gamma band (30-50 Hz), musicians showed robust changes in complexity, consistent over various scalp regions, during listening to music from resting condition, whereas such changes were minimal for non-musicians. Then the index is used to separate healthy participants from epileptic and manic patients based on spontaneous EEG analysis. Finally, it is used to track a tonic-clonic seizure EEG signal, and a conspicuous change was found in the complexity profiles of delta band (1-3.5 Hz) oscillations at the onset of seizure. We conclude that this index would be useful for quantification of a wide range of time series including neural oscillations.
不规则和复杂信号在自然界中无处不在。本文的主要目的是开发一种指数，用于量化此类信号的复杂性，该指数基于通过奇异值分解估计的正交振荡模式强度的分布。该指数首先用模拟混沌和/或随机映射及流进行测试。在神经数据分析中，该指数首先应用于从两组人员（音乐家和非音乐家）在听音乐和静息状态下记录的认知脑电图数据集。在伽马波段（30 - 50赫兹），音乐家在从静息状态转为听音乐时，在复杂性上表现出稳健的变化，在各个头皮区域都是一致的，而非音乐家的此类变化则很小。然后，该指数用于基于自发脑电图分析将健康参与者与癫痫患者和躁狂患者区分开来。最后，它被用于跟踪强直 - 阵挛性癫痫发作的脑电图信号，并且在癫痫发作开始时发现δ波段（1 - 3.5赫兹）振荡的复杂性特征有明显变化。我们得出结论，该指数对于量化包括神经振荡在内的广泛时间序列将是有用的。
