Efficient Strategies for Estimating the Spatial
Coherence of Backscatter

Various properties of an ultrasound echo can be analyzed to characterize its source. The magnitude of the echo is the most commonly used property, but is susceptible to image degradation via noise sources such as reverberation clutter and phase aberration. Recent studies suggest that the spatial coherence of the echo, a measure of the similarity at different points along the wavefront, is a more robust echo property in noisy imaging environments. In particular, short-lag spatial coherence (SLSC) beamforming has demonstrated improvements in image quality in cardiac, liver, and fetal imaging clinical studies. However, the computational cost of spatial coherence estimation is a significant hurdle to its application in the clinic.

In this work, we propose several efficient strategies for spatial coherence estimation based on the statistical properties of ultrasound backscatter. Using these techniques, SLSC images of similar quality can be formed with a 95% reduction in computation time.

Images of the heart were formed with (a) the magnitude of the echo, (b) the original SLSC formulation, and (c-e) successively more computationally efficient SLSC techniques, and the computational throughput of each method is charted in (f). All of the SLSC images (b-e) show reduced clutter near the apex of the heart. By combining the proposed strategies, the image in (e) is formed 20 times faster than the conventional image in (b).

D. Hyun; A. L. C. Crowley; J. J. Dahl, "Efficient Strategies for Estimating the Spatial Coherence of Backscatter," in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control , vol.PP, no.99, pp.1-1 doi: 10.1109/TUFFC.2016.2634004