Lightweight Physics-Aware Zero-Shot Ultrasound Plane-Wave Denoising

TL;DR

This paper introduces a lightweight, physics-aware zero-shot denoising method for ultrasound imaging that enhances image quality without needing external datasets or clean references, using a self-supervised learning approach.

Contribution

The authors propose a novel zero-shot denoising framework that leverages physics-aware pairing of steering angles for ultrasound, eliminating the need for training data or fine-tuning.

Findings

Improves ultrasound image quality without external training data.

Uses a simple two-layer CNN for fast, efficient training.

Effectively distinguishes anatomical structures from noise and artifacts.

Abstract

Ultrasound Coherent Plane-Wave Compounding (CPWC) enhances image contrast by combining echoes from multiple steered transmissions. While increasing the number of steering angles generally improves image quality, it significantly reduces frame rate and may introduce blurring artifacts in fast-moving targets. In addition, compounded images remain susceptible to noise, particularly when acquired using a limited number of transmissions. In this work, we propose a lightweight physics-aware zero-shot denoising framework for low-angle CPWC ultrasound imaging that improves image quality without requiring external training datasets or clean reference images. The proposed approach partitions the available steering angles into two disjoint subsets, each used to reconstruct compounded images with different angle-dependent artifacts and noise characteristics. These reconstructed images are then used as pseudo-pairs within a self-supervised residual learning framework to train a lightweight convolutional neural network directly on the test sample. Because the underlying tissue structures remain consistent across the subsets while the incoherent artifacts vary with steering angle selection, the proposed physics-aware pairing strategy enables the network to distinguish anatomical information from inconsistent noise and artifacts. Unlike supervised approaches, the proposed method does not require domain-specific fine-tuning or paired datasets, making it adaptable across different anatomical regions and acquisition settings. Furthermore, the proposed framework employs an efficient architecture composed of only two convolutional layers, enabling fast and computationally inexpensive training.