3-D Ultrasound Imaging with Microbeamformer-Based FDMAS: a Preliminary Performance Assessment

High-end 3D systems achieve high-quality imaging by using densely populated 2D array probes. To cope with thousands of elements, the classic delay and sum (DAS) beamforming is split into two stages: (1) inside the probe, the micro-beamformer (µB) partially performs DAS on sub-groups of elements; (2) the system completes the DAS beamforming by properly delaying and summing the µB outputs. Although DAS is implemented on most scanners, there is increasing interest for alternative beamforming methods, such as the filtered delay multiply and sum (FDMAS), which was shown effective in improving the 2D image quality with a moderate complexity increase. Therefore, this work assesses the performance of FDMAS in 3D imaging systems using a µB. A 2.5-MHz cardiac probe, with 48×48 elements and a µB handling sub-groups of M×M elements (M=2, 4, 8), was simulated. FDMAS and DAS were distributed between the µB and the system stages according to three different architectures to reconstruct the corresponding point spread functions (PSFs). These were evaluated in terms of -12dB lateral resolution (RES), side-to-main lobe energy ratio (SMER), and highest secondary-lobe-level (SLL). The results show that, compared to the reference DAS implementation, FDMAS realized at the µB stage improves RES but worsens SMER and SLL. However, when implemented on the system, following a standard DAS µB, FDMAS reduces grating lobe levels, improves RES, and achieves comparable SMER and SLL.