Speckle tracking techniques for ultrasound elasticity imaging.

Abstract

The fundamental goal of elasticity imaging is to detect changes in soft tissue elasticity which are related to pathological processes. To reconstruct images of elastic modulus deep within the body, mechanical displacement and strain fields must be measured. Using controlled deformations at the body surface and ultrasound imaging, the motion of the tissue can be estimated using speckle tracking algorithms. Traditional ultrasound tracking methods are examined and new methods are presented which account for the specific needs of elasticity imaging. In ultrasound elasticity imaging, strain decorrelation is a major source of error in displacements estimated using correlation techniques. This error can be significantly decreased by both reducing the correlation integration time and filtering the correlation functions prior to displacement estimation. High resolution, high signal-to-noise ratio (SNR) strain estimates can be computed using small correlation kernels (on the order of the correlation width of the ultrasound signal) and correlation filtering. Since errors in displacement and strain estimates depend on the magnitude of the induced strain, strain SNR will be a function of the applied deformation. By applying continuous deformation and capturing data in real-time, the surface displacement providing the highest strain SNR can be selected retrospectively. A method to adaptively optimize strain SNR over the image plane using retrospective processing is presented. Using the incompressibility property of soft tissue, lateral displacements can be reconstructed from axial strain measurements. Due to anisotropy in spatial frequencies, axial displacements are much more accurate than lateral ones. Incompressibility processing greatly improves the accuracy and SNR of lateral displacement measurements compared with more traditional speckle tracking. To test the accuracy and precision of these algorithms, experimental elasticity imaging systems were constructed to apply controlled deformations and collect ultrasound data. Results from simulations and experiments on gelatin phantoms are presented. The potential application of these methods to diagnose kidney transplant rejection is examined. Initial studies from gelatin phantoms with embedded ex vivo kidneys are presented, in which a cross-linking agent was used to modify kidney elasticity. The results suggest that ultrasound elasticity imaging is possible in complex structures such as the kidney using these speckle tracking techniques.