THE NONSTATIONARY STRAIN FILTER IN ELASTOGRAPHY .1. FREQUENCY-DEPENDENT ATTENUATION

The accuracy and precision of the strain estimates in elastography dep end on a myriad number of factors, A clear understanding of the variou s factors (noise sources) that plague strain estimation is essential t o obtain quality elastograms. The nonstationary variation in the perfo rmance of the strain filter due to frequency-dependent attenuation and lateral and elevational signal decorrelation are analyzed in this and the companion paper for the cross-correlation-based strain estimator, In this paper, we focus on the role of frequency-dependent attenuatio n in the performance of the strain estimator, The reduction in the sig nal-to-noise ratio (SNR,) in the RF signal, and the center frequency a nd bandwidth downshift with frequency-dependent attenuation are incorp orated into the strain filter formulation, Both linear and nonlinear f requency dependence of attenuation are theoretically analyzed, Monte-C arlo simulations are used to corroborate the theoretically predicted r esults. Experimental results illustrate the deterioration in the preci sion of the strain estimates with depth in a uniformly elastic phantom , Theoretical, simulation and experimental results indicate the import ance of high SNR,values in the RF signals, because the strain estimati on sensitivity, elastographic SNR, and dynamic range deteriorate rapid ly with a decrease in the SNR,, In addition, a shift in the strain fil ter toward higher strains is observed at large depths in tissue due to the center frequency downshift. (C) 1997 World Federation for Ultraso und in Medicine & Biology.