Sub-Nyquist Sampling and Fourier Domain Beamforming in Volumetric Ultrasound Imaging

• Published in: IEEE transactions on ultrasonics, ferroelectrics, and frequency control Vol. 63; no. 5; pp. 703 - 716
• Main Authors: Burshtein, Amir, Birk, Michael, Chernyakova, Tanya, Eilam, Alon, Kempinski, Arcady, Eldar, Yonina C.
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.05.2016; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Array Processing; Array signal processing; Arrays; Beamforming; Compressed Sensing; Fourier Analysis; Frequency-domain analysis; Humans; Image Processing, Computer-Assisted - methods; Imaging; Models, Cardiovascular; Phantoms, Imaging; Three-dimensional displays; Transducers; Ultrasonic imaging; Ultrasonography - methods; Ultrasound; compressed sensing (CS); beamforming; Array processing; ultrasound
• ISSN: 0885-3010; 1525-8955
• DOI: 10.1109/TUFFC.2016.2535280

A key step in ultrasound image formation is digital beamforming of signals sampled by several transducer elements placed upon an array. High-resolution digital beamforming introduces the demand for sampling rates significantly higher than the signals' Nyquist rate, which greatly increases the volume of data that must be transmitted from the system's front end. In 3-D ultrasound imaging, 2-D transducer arrays rather than 1-D arrays are used, and more scan lines are needed. This implies that the amount of sampled data is vastly increased with respect to 2-D imaging. In this work, we show that a considerable reduction in data rate can be achieved by applying the ideas of Xampling and frequency domain beamforming (FDBF), leading to a sub-Nyquist sampling rate, which uses only a portion of the bandwidth of the ultrasound signals to reconstruct the image. We extend previous work on FDBF for 2-D ultrasound imaging to accommodate the geometry imposed by volumetric scanning and a 2-D grid of transducer elements. High image quality from low-rate samples is demonstrated by simulation of a phantom image composed of several small reflectors. Our technique is then applied to raw data of a heart ventricle phantom obtained by a commercial 3-D ultrasound system. We show that by performing 3-D beamforming in the frequency domain, sub-Nyquist sampling and low processing rate are achievable, while maintaining adequate image quality.