Simultaneous Reverberation Noise Reduction and Aberration Correction Using Wavefield Correlation

Reverberation and aberration contribute to in vivo ultrasound image clutter, affecting clinical diagnoses especially in areas such as abdominal imaging. Existing clutter reduction techniques typically target only one source of image degradation, and recent techniques for local sound speed estimation and subsequent aberration correction can have degraded performance with reverberation noise present in the data. However, beamforming using wavefield correlation allows for reduction of both sources of clutter in the final image due to its correlation of propagated transmit and receive wavefields, which results in spatiotemporal matched filtering of the data. In simulations containing both aberration and reverberation from a 10 mm abdominal wall layer with varied tissue structures, wavefield correlation using a local sound speed map results in noticeably higher contrast anechoic cyst regions, with GCNR values consistently higher by up to 40+% when compared to standard delay and sum (DAS) beamformers. In addition, cyst boundaries appear sharper in wavefield correlation images. For in vivo thyroid data where the ground truth sound speed was not known, the reverberation reduction effect of constant sound speed wavefield correlation was still noticeable, with lesion GCNRs typically being 2-3% higher than for constant sound speed DAS. These results show that wavefield correlation can reduce multiple sources of clutter simultaneously, offering promising opportunities for clinical image quality improvements in difficult-to-image patients.