| The presented work examines the two extremes of ultrasound beamforming---optimal beamforming for the best attainable image quality on one hand, and low-cost beamforming for simple and inexpensive implementation on the other.; A non-iterative algorithm for the optimal design of aperture weights, the minimum sum squared error (MSSE) technique, was developed. The MSSE technique enables the design of arbitrary beam patterns in continuous wave and broadband systems. Simulations and experiments indicated that the MSSE technique can be used to obtain a wide range of system responses.; An extremely simple and compact beamforming scheme, named direct sampled I/Q (DSIQ) beamforming, was also developed for low-cost systems. DSIQ beamforming implements focusing by phase rotation of in-phase/quadrature (I/Q) data that are obtained by directly sampling the received radio frequency echo signal. Simulations and experiments indicated that DSIQ beamforming yields adequate image quality despite being a narrowband focusing scheme.; Finally, a novel contrast resolution metric was used to characterize the performance of a 2D array based low-cost system, called the Sonic Window, under development at the University of Virginia that implements DSIQ beamforming. The metric computes the contrast of an anechoic cyst in a speckle generating background in terms of the system point spread function, accounting for the effects of electronic noise; it enables straightforward estimation of the resolution of ultrasound systems and parameter optimization. The metric was used to optimize system parameters in the current prototype system, including electronic noise, quantization, f/#, apodization, and crosstalk. An electronic signal to noise ratio of 0 dB per channel and 10 bits per real sample were found to yield adequate performance. The Nuttall window was determined to be the best apodization window among the windows that were evaluated. Acoustic and electrical crosstalk were found to have negligible effect on the prototype system.; Finally, the impacts of spatial and frequency compounding on lesion detectability in the prototype system were theoretically studied using simplified k-space representations of the system response. The analysis suggested that spatial and frequency compounding increase lesion detectability by up to 212% and 376% respectively. |
