Ultrasound Autofocusing: Common Midpoint Phase Error Optimization via Differentiable Beamforming.

Area of Science:

• Medical Imaging
• Acoustics
• Wave Propagation

Background:

• Ultrasound imaging suffers from phase aberrations caused by heterogeneous media, degrading image resolution and contrast.
• Adaptive imaging techniques aim to correct these aberrations for improved focusing and image quality.

Purpose of the Study:

• To propose and validate a novel autofocusing paradigm for aberration correction in ultrasound imaging.
• To utilize common midpoint phase error (CMPE) as a robust measure for phase aberration quantification and correction.

Main Methods:

• Fitting an acoustic velocity field to pressure measurements by optimizing CMPE.
• Employing a straight-ray wave propagation model and differentiable beamforming for iterative focus improvement and velocity field estimation.
• Utilizing wavefield measurements and a straight-ray integral solution of two-way time-of-flight, avoiding explicit time-stepping models.

Main Results:

• Common midpoint phase error (CMPE) is demonstrated as a reliable metric for phase aberration in heterogeneous media.
• The proposed method successfully improves image focus while simultaneously estimating the acoustic velocity field.
• Successful validation across in silico simulations, in vitro phantoms, and in vivo mammalian models.

Conclusions:

• The developed acoustic autofocusing paradigm effectively corrects phase aberrations in ultrasound imaging.
• The method enables distributed aberration quantification, correction, and velocity estimation for enhanced medical ultrasound applications.
• This approach offers a practical solution for improving ultrasound image quality without complex wave propagation modeling.