Hoffmann U, Yanar A, Franzeck U K, Edwards J M, Bollinger A
Department of Internal Medicine, University Hospital, Zurich, Switzerland.
Microvasc Res. 1990 Nov;40(3):293-301. doi: 10.1016/0026-2862(90)90028-p.
A digital filter system (DFS) was implemented to describe the frequency components of laser Doppler signals. The DFS calculates a frequency histogram on the basis of the prevalence of a certain frequency band during the sample period. This approach seems to be more suitable for the detection of low-power small-amplitude flux motion components recorded by laser Doppler as compared to the commonly used fast Fourier algorithm. Fourier analysis provides an amplitude-relevant frequency spectrum which may overlook relevant frequency components with small amplitude due to low power or short duration. DFS allows better detection of such signals as demonstrated by the initial clinical results of this study evaluating flux motion on the feet of 4 healthy controls and 10 patients with peripheral arterial occlusive disease. Three frequency components are easily distinguished corresponding to low (3.6 +/- 1.5 cycles/minute) and high (18.6 +/- 4.5 cycles/min) frequency flux motion and the pulsatile waves due to cardiac action.
采用数字滤波系统（DFS）来描述激光多普勒信号的频率成分。DFS根据采样期内特定频带的出现情况计算频率直方图。与常用的快速傅里叶算法相比，这种方法似乎更适合检测激光多普勒记录的低功率小幅度通量运动成分。傅里叶分析提供了一个与幅度相关的频谱，由于功率低或持续时间短，可能会忽略小幅度的相关频率成分。本研究对4名健康对照者和10名外周动脉闭塞性疾病患者足部的通量运动进行了初步临床评估，结果表明DFS能更好地检测此类信号。对应于低频（3.6±1.5次/分钟）和高频（18.6±4.5次/分钟）通量运动以及心脏活动引起的脉搏波，很容易区分出三种频率成分。
