Mitigating Aberration-Induced Noise: A Deep Learning-Based Aberration-to-Aberration Approach

TL;DR

This paper introduces a novel deep learning approach for ultrasound phase aberration correction that trains directly on real aberrated data without ground truth, using an adaptive loss function and a large publicly available dataset.

Contribution

It presents the first ground truth-free deep learning method for phase aberration correction in ultrasound imaging, utilizing a mixed loss function and real data training.

Findings

Effective correction of aberrations demonstrated

Adaptive loss function improves convergence

Large dataset supports future research

Abstract

One of the primary sources of suboptimal image quality in ultrasound imaging is phase aberration. It is caused by spatial changes in sound speed over a heterogeneous medium, which disturbs the transmitted waves and prevents coherent summation of echo signals. Obtaining non-aberrated ground truths in real-world scenarios can be extremely challenging, if not impossible. This challenge hinders the performance of deep learning-based techniques due to the domain shift between simulated and experimental data. Here, for the first time, we propose a deep learning-based method that does not require ground truth to correct the phase aberration problem and, as such, can be directly trained on real data. We train a network wherein both the input and target output are randomly aberrated radio frequency (RF) data. Moreover, we demonstrate that a conventional loss function such as mean square error is inadequate for training such a network to achieve optimal performance. Instead, we propose an adaptive mixed loss function that employs both B-mode and RF data, resulting in more efficient convergence and enhanced performance. Finally, we publicly release our dataset, comprising over 180,000 aberrated single plane-wave images (RF data), wherein phase aberrations are modeled as near-field phase screens. Although not utilized in the proposed method, each aberrated image is paired with its corresponding aberration profile and the non-aberrated version, aiming to mitigate the data scarcity problem in developing deep learning-based techniques for phase aberration correction.