A Sparse regularization approach for ultrafast ultrasound imaging

Rafael Carrillo (1, 2) , Adrien Besson (2) , Zhang Miaomiao (3) , Denis Friboulet (3) , Yves Wiaux (4) , Jean-Philippe Thiran (2, 1) , Olivier Bernard (3)

1 Departments of Radiology
2 LTS5 - Signal Processing Laboratory [Lausanne]
3 Images et Modèles
4 HWU - Heriot-Watt University [Edinburgh]

Ultrafast imaging based on plane-wave (PW) insonification is an active area of research due to its capability of reaching high frame rates. Several approaches have been proposed either based on either of Fourier-domain reconstruction or on delay-and-sum (DAS) reconstruction. Using a single PW, these techniques achieve low quality, in terms of resolution and contrast, compared to the classic DAS method with focused beams. To overcome this drawback, compounding of several steered PWs is needed, which currently decreases the high frame rate limit that could be reached by such techniques. Based on a compressed sensing (CS) framework, we propose a new method that allows the reconstruction of high quality ultrasound (US) images from only 1 PW at the expense of augmenting the computational complexity at the reconstruction.

Domaines: Traitement du signal et de l'image [eess.SP]; Acoustique [physics.class-ph]

Rafael Carrillo, Adrien Besson, Zhang Miaomiao, Denis Friboulet, Yves Wiaux, et al.. A Sparse regularization approach for ultrafast ultrasound imaging. Ultrasonics Symposium (IUS), 2015 IEEE International, Oct 2015, Taipei, Taiwan. pp.1-4, ⟨10.1109/ULTSYM.2015.0184⟩. ⟨hal-01272389⟩