Real time 3D US-tagging combined with 3D phase-based motion estimation

Sébastien Salles (1) , H. Liebgott (1) , Damien Garcia (2) , D. Vray (1)

1 Imagerie Ultrasonore
2 CR CHUM - Centre de Recherche du Centre Hospitalier de l’Université de Montréal

By contrast with 2D imaging, quantitative analysis of 3D motion from ultrasound images can provide improved information in several applications, such as arterial mechanical assessment, heart motion and blood flow. Unfortunately, it remains difficult to obtain a high definition of the motion estimate in the lateral and elevation directions (i.e. perpendicular to the beam axis). To increase the definition in both these directions, this paper presents a 3D extension of a the transverse oscillations method that enables one to obtain ultrasound fields featuring oscillations along the 3 spatial dimensions, using a single apodization function. The 3D motion method is estimated using the phases of the images. Simulation results show that a 3D trajectory can be followed with a relative mean error smaller than 8%.

Mots clés: Ultrasonic imaging; Oscillators; Motion estimation; Estimation; Trajectory; Imaging; Three-dimensional displays

Domaines: Imagerie

Sébastien Salles, H. Liebgott, Damien Garcia, D. Vray. Real time 3D US-tagging combined with 3D phase-based motion estimation. 2013 IEEE International Ultrasonics Symposium (IUS), Jul 2013, Prague, Czech Republic. pp.585-588, ⟨10.1109/ULTSYM.2013.0151⟩. ⟨hal-00850061⟩