A new method for Doppler frequency analysis that promises a major improvement in performance

The Doppler grey scale waveform obtained by frequency analysis is related to the blood flow velocity in the vessel under examination. Consequently, detailed evaluation of the Doppler spectral waveform or the instantaneous Doppler spectrum is clinically useful for assessing arterial disease. The purpose of this paper is to demonstrate the potential advantages of a new frequency analysis method (autoregressive moving average) for processing and displaying continuous wave and pulsed Doppler signals by comparing the quality of the Doppler spectral waveforms produced by this new method to those obtained by standard Fast Fourier transform analysis. Continuous wave pulsed Doppler recordings were stored on a computer and reprocessed by Fast Fourier transform and autoregressive moving average methods. Varying degrees of noise were added to both types of Doppler recordings in order to simulate the problems encountered in the clinical vascular laboratory. By comparing the two methods using standard Doppler spectral waveforms and 3-dimensional spectral plots, under adverse conditions of significant noise, the autoregressive moving average technique proved to be superior to the classical Fast Fourier approach. It is concluded that the autoregressive method of frequency analysis offers the potential for achieving significant improvements in the display of the Doppler spectral waveform when compared to the classical Fast Fourier approach. Such improvements may improve the accuracy in arterial disease assessment.