Twofold minimum variance beamforming for enhanced ultrasound imaging

Abstract

Purpose

To calculate the optimal weights of a minimum variance (MV) beamformer, the covariance matrix of the received data should be estimated. In ultrasound imaging, it is done through spatial smoothing. However, this technique reduces the effective aperture and consequently limits the attainable resolution. To mitigate this effect, we propose a new configuration for applying an adaptive beamforming method.

Methods

The method is based on applying MV in a twofold configuration such that the outputs of the subarrays are combined through an adaptive MV-like beamformer, instead of simply summing them. In this way, a fully adaptive beamformer is obtained.

Results

Applying the method on a phantom containing point targets has shown that off-axis signals are attenuated more in comparison to MV. This results in a lowered mainlobe width and decreased sidelobe levels. Moreover, simulation results of a cyst phantom confirm the superiority of the method over the MV in terms of contrast.

Conclusion

The method yields significant improvement in the resolving capability and the contrast compared with the conventional MV. At the presence of the steering vector errors, these superiorities were achieved at a cost of slightly more errors in estimating the reflectivity coefficients.