A Deep Reinforcement Learning Based Region-Specific Beamformer for Sparse Arrays 3-D Ultrasound Imaging
IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control · 2025
DOI: 10.1109/tuffc.2025.3560872 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 2025. Verfasst von Mohamed Tamraoui und 2 weiteren Autor:innen. Hauptthemen laut MEDLINE: Imaging, Three-Dimensional, Deep Learning. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 72 Literaturangaben. Gefördert von: LABEX PRIMES, Université de Lyon through the Program “In- vestissements d’Avenir”, French National Research Agency (ANR) and performed within the framework.
Abstract (Originalsprache)
Sparse arrays offer several advantages over other element reduction techniques for 3-D ultrasound imaging. However, the large interelement spacing in these arrays results in high sidelobe-related artifacts, which significantly degrade image quality and limit their application in 3-D ultrasound imaging. Adaptive beamformers have been proposed to mitigate sidelobe-related artifacts, but they often degrade speckle texture quality, resulting in unnaturally dark images. To overcome these limitations, we propose RSB-Net, a region-specific beamformer based on deep reinforcement learning (DRL). RSB-Net adaptively selects the most suitable beamformer for each pixel of the image, applying adaptive beamforming in regions dominated by sidelobe artifacts and delay-and-sum (DAS) beamforming in regions where speckle texture should be preserved. The effectiveness of RSB-Net was validated on both simulated and experimental synthetic transmit aperture (STA) RF datasets with a newly designed sparse array prototype. On simulated data, RSB-Net achieved significant gains, with improvements of 52.81 dB in contrast ratio (CR) and 0.65 in a generalized contrast-to-noise ratio (gCNR) compared to DAS beamforming. In experimental tissue-mimicking phantom data, RSB-Net demonstrated similar performance, achieving gains of 51.01 dB and 0.64, respectively. These results highlight the potential of RSB-Net as a robust and effective solution for high-quality B-mode 3-D ultrasound imaging using 2-D sparse arrays, advancing the standardization of 3-D ultrasound in clinical settings by enhancing anatomical visualization, reducing operator dependency, and improving measurement accuracy for lesions and calcifications.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Förderung
- LABEX PRIMES
- Université de Lyon through the Program “In- vestissements d’Avenir”
- French National Research Agency (ANR) and performed within the framework
Zitieren
APA
Tamraoui, M., Liebgott, H., & Roux, E. (2025). A Deep Reinforcement Learning Based Region-Specific Beamformer for Sparse Arrays 3-D Ultrasound Imaging. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control. https://doi.org/10.1109/tuffc.2025.3560872
Vancouver
Tamraoui M, Liebgott H, Roux E. A Deep Reinforcement Learning Based Region-Specific Beamformer for Sparse Arrays 3-D Ultrasound Imaging. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control. 2025. doi:10.1109/tuffc.2025.3560872
BibTeX
@article{Tamraoui2025,
author = {Tamraoui, Mohamed and Liebgott, Herve and Roux, Emmanuel},
title = {A Deep Reinforcement Learning Based Region-Specific Beamformer for Sparse Arrays 3-D Ultrasound Imaging},
journal = {IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control},
year = {2025},
doi = {10.1109/tuffc.2025.3560872}
}
EUVIMED