A novel spline-based algorithm for multidimensional displacement and strain estimation
In medical ultrasound, motion estimation is used to align images in extended field of view applications, estimate blood or tissue motion, and estimate radiation force or mechanically induced displacements for elasticity imaging. In all of these applications the sampled nature of real-world images makes precise estimation of sub-sample displacements computationally costly at best, and impossible at worst. In this paper we describe a novel MUlti-dimensional Spline-based motion Estimator (MUSE). We performed simulations and experiments to assess the intrinsic bias and standard deviation of this algorithm for speckle kernels 4 samples wide and 16 samples deep. In 1000 noise-free simulations we found that MUSE exhibits maximum bias errors of 0.002 and 0.0003 samples (0.040 μm and 0.045 μm) in range and azimuth respectively. The maximum simulated standard deviation of estimates in both dimensions was comparable at 0.0026 samples (0.05 μm in range and 0.78 μm in azimuth). 25,600 motion estimates were also performed using experimental data acquired using an Ultrasonix Sonix RP imaging system with a L14-5/38 linear array transducer operated at 6.6 MHz. With this experimental data we found that bias errors were apparently significantly smaller than geometric errors induced by machining of the transducer mount. These simulated and experimental results significantly outperform other published algorithms. Straightforward extension to include companding and shear estimation make this algorithm particularly attractive for tissue elasticity imaging applications. © 2006 IEEE.
