High-Resolution Bone Ultrasound Tomography via Multi-Angle Plane-Wave Full Waveform Inversion

Full waveform inversion (FWI) has shown great potential for high resolution ultrasound imaging by utilizing complete wavefield information. However, conventional FWI approaches are associated with high computational costs, whereas ray-based methods often fail to accurately capture complex micro bone structure. In this study, we propose a full waveform ultrasound computed tomography (USCT) method based on multi-angle plane-wave FWI to enable efficient and high resolution reconstruction of sound speed map in bone imaging. By employing the angular spectrum method (ASM) for forward modeling and adjoint-state back-propagation for gradient computation, the inverse problem is solved using a damped conjugate gradient algorithm to minimize data misfit across multiple angles and frequencies, while effectively mitigating cycle-skipping issues. Numerical experiments conducted with 180 plane-wave angles successfully reconstructed detailed bone structures, demonstrating enhanced resolution and improved robustness compared to ray-based approaches. The proposed framework provides a computationally efficient solution for high-resolution USCT, with future work focusing on validation through physical and ex-vivo bone imaging experiments.