2-D companding for noise reduction in strain imaging

摘要

Companding is a signal preprocessing technique for improving thenprecision of correlation-based time delay measurements. In strainnimaging, companding is applied to warp 2-D or 3-D ultrasonic echo fieldsnto improve coherence between data acquired before and after compression.nIt minimizes decorrelation errors, which are the dominant source ofnstrain image noise. The word refers to a spatially variable signalnscaling that compresses and expands waveforms acquired in an ultrasonicnscan plane or volume. Temporal stretching by the applied strain is ansingle-scale (global), 1-D companding process that has been usednsuccessfully to reduce strain noise. This paper describes a two-scalen(global and local), 2-D companding technique that is based on ansum-absolute-difference (SAD) algorithm for blood velocity estimation.nSeveral experiments are presented that demonstrate improvements inntarget visibility for strain imaging. The results show that, if tissuenmotion can be confined to the scan plane of a linear array transducer,ndisplacement variance can be reduced two orders of magnitude using 2-Dnlocal companding relative to temporal stretching