IRSDE-Despeckle: A Physics-Grounded Diffusion Model for Generalizable Ultrasound Despeckling

TL;DR

This paper introduces a physics-grounded diffusion model for ultrasound despeckling that outperforms classical and learning-based methods, providing uncertainty quantification and insights into domain shift for better clinical robustness.

Contribution

We develop a diffusion-based ultrasound despeckling method grounded in physical modeling, with a large simulated dataset and uncertainty estimation for improved clinical applicability.

Findings

Outperforms classical filters and recent learning-based methods on simulated data

Provides uncertainty estimates that correlate with reconstruction errors

Identifies domain shift issues related to simulation parameters

Abstract

Ultrasound imaging is widely used for real-time, noninvasive diagnosis, but speckle and related artifacts reduce image quality and can hinder interpretation. We present a diffusion-based ultrasound despeckling method built on the Image Restoration Stochastic Differential Equations framework. To enable supervised training, we curate large paired datasets by simulating ultrasound images from speckle-free magnetic resonance images using the Matlab UltraSound Toolbox. The proposed model reconstructs speckle-suppressed images while preserving anatomically meaningful edges and contrast. On a held-out simulated test set, our approach consistently outperforms classical filters and recent learning-based despeckling baselines. We quantify prediction uncertainty via cross-model variance and show that higher uncertainty correlates with higher reconstruction error, providing a practical indicator of difficult or failure-prone regions. Finally, we evaluate sensitivity to simulation probe settings and observe domain shift, motivating diversified training and adaptation for robust clinical deployment.