Minimum Variance adaptive beamforming for ultrasound imaging

The conventional Delay-And-Sum (DAS) beamformer is the standard method to process the signals received by an array of sensors in ultrasound imaging. In DAS beamforming, the sensor signals are time-delayed, weighted, and subsequently summed to form a maximized beamformer output. Adaptive beamforming results in a higher image quality both in terms of resolution and contrast, when compared to the conventional one. One of the most common adaptive methods is the Minimum Variance (MV) beamformer. Similar to conventional processing is required with the only difference that the apodization weights do not have fixed values but are data-dependent. The objective of MV beamformer is to find those apodization weights that will attenuate all signal and noise from the beamformer output as much as possible, while keeping signals of interest without any distortion. The method is implemented in Frequency domain and is validated with simulated ultrasound data. Computer phantoms including point targets and circular cysts have been used and the Full-Width-Half-Maximum (FWHM), the Peak-Side lobe-Level (PSL), and the contrast level have been adopted as criteria for the comparison. Resolution is more than 20 times higher in the case of minimum variance adaptive beamformer but comes at the expense of much higher computational time. This is the major limitation for using the MV adaptive beamformer for real time applications, where the complexity of the beamformer should be kept at the lowest possible level. An alternative implementation of the same beamformer in the time domain will also be discussed as this way it is possible to maintain the resolution almost at the same levels but at the same time to reduce the computational burden significantly.