DiffSOS: Acoustic Conditional Diffusion Model for Speed-of-Sound Reconstruction in Ultrasound Computed Tomography

TL;DR

DiffSOS introduces a novel conditional diffusion model for ultrasound speed-of-sound reconstruction, achieving high accuracy, near real-time inference, and uncertainty estimation, thus advancing clinical imaging capabilities.

Contribution

The paper presents DiffSOS, a diffusion-based framework with a specialized ControlNet and hybrid loss for accurate, fast, and uncertainty-aware SoS mapping from acoustic waveforms.

Findings

Outperforms state-of-the-art networks on OpenPros USCT benchmark

Achieves an average Multi-scale Structural Similarity of 0.957

Enables near real-time reconstruction with 10 DDIM steps

Abstract

Accurate Speed-of-Sound (SoS) reconstruction from acoustic waveforms is a cornerstone of ultrasound computed tomography (USCT), enabling quantitative velocity mapping that reveals subtle anatomical details and pathological variations often invisible in conventional imaging. However, practical utility is hindered by the limitations of existing algorithms; traditional Full Waveform Inversion (FWI) is computationally intensive, while current deep learning approaches tend to produce oversmoothed results lacking fine details. We propose DiffSOS, a conditional diffusion model that directly maps acoustic waveforms to SoS maps. Our framework employs a specialized acoustic ControlNet to strictly ground the denoising process in physical wave measurements. To ensure structural consistency, we optimize a hybrid loss function that integrates noise prediction, spatial reconstruction, and noise frequency content. To accelerate inference, we employ stochastic Denoising Diffusion Implicit Model (DDIM) sampling, achieving near real-time reconstruction with only 10 steps. Crucially, we exploit the stochastic generative nature of our framework to estimate pixel-wise uncertainty, providing a measure of reliability that is often absent in deterministic approaches. Evaluated on the OpenPros USCT benchmark, DiffSOS significantly outperforms state-of-the-art networks, achieving an average Multi-scale Structural Similarity of 0.957. Our approach provides high-fidelity SoS maps with a principled measure of confidence, facilitating safer and faster clinical interpretation.