Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Yujia Wu, Shuoqi Chen, Shiru Wang, Yucheng Tang, Petr Bruza, Geoffrey P. Luke

TL;DR

This work proposes DiffSOS, a conditional diffusion model that directly maps acoustic waveforms to SoS maps with a principled measure of confidence, facilitating safer and faster clinical interpretation and exploiting the stochastic generative nature of the framework to estimate pixel-wise uncertainty.

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.

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

TL;DR

This work proposes DiffSOS, a conditional diffusion model that directly maps acoustic waveforms to SoS maps with a principled measure of confidence, facilitating safer and faster clinical interpretation and exploiting the stochastic generative nature of the framework to estimate pixel-wise uncertainty.

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.
Paper Structure (14 sections, 8 equations, 4 figures, 2 tables)

This paper contains 14 sections, 8 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Proposed Method
  3. Acoustic ControlNet:
  4. Hybrid Multi-objective Loss:
  5. Stochastic Inference and Uncertainty:
  6. Experiments and Results
  7. Dataset.
  8. Experimental Setup
  9. Results
  10. Model Comparison
  11. Ablation Studies
  12. Uncertainty
  13. Efficiency
  14. Discussion and Conclusion

Figures (4)

  • Figure 1: Overview of the proposed DiffSOS framework.
  • Figure 2: Qualitative comparison with baseline methods.
  • Figure 3: Visual ablation study of loss components.
  • Figure 4: Uncertainty and Efficiency Analysis.(a) Uncertainty maps highly correlate with reconstruction errors. Quantitative trade-offs for ensemble size (b) and sampling steps (c) confirm that high-fidelity reconstruction is achievable within clinically viable timeframes.