Phase Aberration Correction without Reference Data: An Adaptive Mixed Loss Deep Learning Approach
This addresses a key challenge in ultrasound image quality for medical diagnostics by enabling correction without ground truths, though it is incremental as it builds on existing deep learning approaches.
The paper tackles phase aberration correction in ultrasound imaging without needing reference data by proposing a deep learning method trained on randomly aberrated RF data and an adaptive mixed loss function, achieving enhanced performance and more efficient convergence.
Phase aberration is one of the primary sources of image quality degradation in ultrasound, which is induced by spatial variations in sound speed across the heterogeneous medium. This effect disrupts transmitted waves and prevents coherent summation of echo signals, resulting in suboptimal image quality. In real experiments, obtaining non-aberrated ground truths can be extremely challenging, if not infeasible. It hinders the performance of deep learning-based phase aberration correction techniques due to sole reliance on simulated data and the presence of domain shift between simulated and experimental data. Here, for the first time, we propose a deep learning-based method that does not require reference data to compensate for the phase aberration effect. We train a network wherein both 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 the 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. Source code is available at \url{http://code.sonography.ai}.