IMPROVED CHARACTERIZATION OF NONSTATIONARY FLOWS USING A REGULARIZED SPECTRAL-ANALYSIS OF ULTRASOUND DOPPLER SIGNALS

This paper addresses the problem of ultrasound Doppler spectral estima tion when only a short observation set is available. Following the wor k of Kitagawa and Gersch, the spectra are described by a long autoregr essive model whose coefficients are estimated in a Bayesian regularize d least squares framework accounting for spectral smoothness in order to avoid too spiky spectra. The critical computation of the tradeoff p arameters is addressed using both maximum likelihood and generalized c ross validation criteria in order to automatically tune the smoothness constraint. The practical potential of the method is demonstrated usi ng both simulated and in vitro signals. In a Monte-Carlo simulation st udy, investigation of quantitative indices such as quadratic distances shows interesting improvements with respect to the usual least square s method whatever the window data length and the signal to noise ratio . When applied to actual Doppler signals, the proposed method offers b etter description of the Doppler spectrum morphology than the usual le ast squares one.