A Least-Squares Strain Estimator for Elastography

TLDR

The authors propose a least‑squares strain estimator (LSQSE) for elastography. The LSQSE is theoretically illustrated with a modified strain filter and experimentally validated on a homogeneous gel phantom. The LSQSE markedly improves elastogram SNR, enhances strain sensitivity and dynamic range, yet introduces a trade‑off between SNR gains and reductions in strain contrast and spatial resolution.

Abstract

A least-squares strain estimator (LSQSE) for elastography is proposed. It is shown that with such an estimator, the signal-to-noise ratio in an elastogram ( SNR e ) is significantly improved. This improvement is illustrated theoretically using a modified strain filter and experimentally using a homogeneous gel phantom. It is demonstrated that the LSQSE results in an increase of the elastographic sensitivity (smallest strain that could be detected), thereby increasing the strain dynamic range. Using simulated data, it is shown that a tradeoff exists between the improvement in SNR e and the reduction of strain contrast and spatial resolution.