Minimizing strain error for in vivo ultra-sound elasticity imaging of human tendon

Ultrasound elasticity imaging (UEI) is a noninvasive method for characterizing mechanical properties of soft tissue. This technique potentially provides important feedback for making treatment decision for advanced tendinopathies, such as posterior tibial tendon dysfunction (PTTD). Highly variable lateral strain along the length of the tendon has been observed during human experiments. To better understand the in vivo results, a dynamic ultrasound model of the PTT under cyclical loading was developed to analyze factors that affect tracking error (e.g., strain magnitude and out-of-plane motion) and compare different methods for calculating strain. The simulations demonstrated that linear fitting of the lateral displacement along the length of the tendon ('Method A') produced less error for displacement and strain calculations compared to the median of the displacement gradients over the region-of-interest ('Method B'). Also, when out-of-plane motion was less than one acoustic wavelength, the strain error was 5.0 ± 4.2% for Method A and 8.3 ± 8.3% for Method B. Out-of-plane motion greater than one acoustic wavelength, produced large strain errors using either method. These results suggest that a linear fit of the displacements along the imaged portion of the tendon is an overall better approach for estimating lateral strain. Optimizing the imaging protocol and data analysis for UEI will lead to a more accurate and quantitative estimate of the mechanical properties of human tendon to help with the diagnosis, staging and treatment decisions for tendinopathies.