Shaswary_Elyas.pdf (43.76 MB)
Download file

Frequency-Domain Synthetic Aperture Focusing Techniques for Imaging with Single-Element Focused Transducers

Download (43.76 MB)
posted on 2021-05-22, 17:00 authored by Elyas Shaswary

Synthetic aperture focusing techniques (SAFT) make the lateral spatial resolution of the conventional ultrasound imaging from a single-element focused transducer more uniform. In this work, two new frequency-domain SAFT (FD-SAFT) algorithms are proposed, which are based on 2D matched filtering techniques. The first algorithm is the FD-SAFT virtual disk source (FD-VDS) that treats the focus of a focused transducer as a finite sized virtual source and the diffraction effect in the far-field is accounted for in the image reconstruction. The second algorithm is the FD-SAFT deconvolution (FD-DC) that uses the simulated point spread function of the imaging system as a matched filter kernel in the image reconstruction. These algorithms were implemented for pulsed and linear frequency modulated chirp excitations. The performance of these algorithms was studied using a series of simulations and experiments, and it was compared with the conventional B-mode and time-domain virtual point source SAFT (TD-VPS) imaging techniques. The image quality was analyzed in terms of spatial resolution, sidelobe level, signal-to-noise ratio (SNR), contrast resolution, contrast-to- speckle ratio, and ex vivo tissue image quality. The results showed that the FD-VDS had the highest spatial resolution and FD-DC had the second highest spatial resolution. In addition, FD-DC had generally the highest SNR. The computation run time of the proposed methods was significantly lower than the TD-VPS. Furthermore, chirp excitation improves the SNR of all methods by about 8 dB without significantly affecting the spatial resolution and sidelobe level. Thus, the FD-VDS and FD-DC methods offer efficient solutions to make the spatial resolution of conventional B-mode imaging more uniform.





Doctor of Philosophy


Biomedical Physics

Granting Institution

Ryerson University

LAC Thesis Type