Hara K, Nakatani S, Ozaki K, Ikeda T, Mogami H
SHARP Corporation, Information Systems Group, Nara, Japan.
Med Inform (Lond). 1990 Apr-Jun;15(2):125-31. doi: 10.3109/14639239008997664.
Intracranial pressure (ICP) oscillation consists of a cardiac-induced component, a respiration-induced component and fluctuation of the base level of ICP. Lundberg reported three types of fluctuations of the base level of ICP with increasing ICP which were referred to as A, B and C waves. Computer algorithms for sampling, processing and displaying ICP data were investigated to depict the power spectrum of ICP oscillations by fast Fourier transform (FFT), thus enabling the B wave to be automatically detected. A power peak was found in the ICP power spectrum between 30 and 120 s, which corresponds to the frequency of the B wave. The maximum power, corresponding to the B-wave amplitude was above 0 dB. An appropriate sampling interval for FFT inputs was about 8 s for real-time processing of the ICP data. The mean ICP value was found useful for making the B-wave peak clearer by reducing the cardiac and respiratory components of ICP oscillations; the window function had no effect on B-wave detection in the ICP power spectrum.
颅内压(ICP)振荡由心脏诱导成分、呼吸诱导成分以及ICP基础水平的波动组成。伦德伯格报告了随着ICP升高,ICP基础水平的三种波动类型,分别称为A波、B波和C波。研究了用于采样、处理和显示ICP数据的计算机算法,以通过快速傅里叶变换(FFT)描绘ICP振荡的功率谱,从而能够自动检测B波。在ICP功率谱中,30至120秒之间发现了一个功率峰值,这与B波的频率相对应。对应于B波振幅的最大功率高于0 dB。对于ICP数据的实时处理,FFT输入的合适采样间隔约为8秒。发现平均ICP值通过减少ICP振荡的心脏和呼吸成分,有助于使B波峰值更清晰;窗函数对ICP功率谱中的B波检测没有影响。