Estimation of quantitative ultrasound parameters derived from backscatter coefficients using plane wave compounding - A comparative simulation study

Abstract

Quantitative ultrasound (QUS) based on backscatter coefficient (BCS) estimation has shown potential for tissue characterization. Even though beamforming using plane wave compounding (PWC) has shown advantages for echographic, Doppler, and elastographic imaging, to this date PWC has not been evaluated for the purpose of BSC estimation. The objective of this study is to analyze through simulations the performance of BSC-based QUS when using PWC versus single focus beamforming. Simulations were performed with FIELD II using a 6.5 MHz, 128-element array with 0.3 mm element pitch. Beamforming was produced both with a single focus at 3 cm and with PWC with 61 transmission angles, with the focal number set to two in both cases. BSCs were estimated using the reference phantom method. The reference and sample phantoms had a concentration of 16 scatterers/mm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> with backscatter cross-sections corresponding to fluid spheres with diameters of 35 and 70 μm, respectively. Both the effective scatterer diameter (ESD) and concentration (ESC) were derived from the estimated BSCs. The effects of noise and mismatches Δc in the sound speed of sample and reference phantoms were assessed by (1) corrupting the pre-beamformed data with additive Gaussian noise corresponding to peak echo signal-to-noise ratios (eSNRs) of down to 10 dB in the single focus beamformed data, and (2) simulating sample phantoms with Δc between -4% and 4%. In the presence of noise, the use of PWC allowed producing estimates with a lower, more uniform variance than the use of a single focus (i.e., PWC ESD and EAC standard deviations of less than 20% and 3 dB, respectively versus single focus ESD and EAC standard deviations of more than 40% and 20 dB, respectively for a peak eSNR of 20 dB). In the presence of phantom sound speed mismatch, both beamforming techniques produced ESD estimates with comparable bias magnitude. Even though PWC beamforming produced smoother ESC error curves, both techniques provided up to 5 dB bias in ESC estimation and outperformed each other under different conditions. These results suggest that PWC may provide benefits to BSC-based QUS imaging.