Reverberation Suppression in Medical Ultrasound Imaging with SURF Processing for Partial Plane Reflectors
MetadataVis full innførsel
Reverberation noise suppression is a major challenge in diagnostic ultrasound imaging to produce reliable images. Common methods for canceling reverberation noise are proven to be effective only for the noise caused by superficial structures, while in medical ultrasound imaging the source of reverberation noise might be located in deeper levels. A newly introduced imaging method (SURF) aims at addressing such a shortcoming. This thesis investigates the effectiveness of SURF in suppressing both superficial and deep reverberation noise. In the first step the method performance is studied for ideal plane reflectors in different depths and it is shown through simulation that an improvement of 19-27dB is expected for the region of interest. Then, a similar study is done for size-reduced reflectors and the results depicts that in case of using dynamic receive focus, reducing the reflector size down to 4 times the beam-width at focal plane does not result in significant performance degradation. Further size reduction degrades the noise suppression at some points in dynamic receive focus case. In the second step, synthetic aperture focusing is utilized to improve the performance for smaller reflectors. This idea is realized by using equally focused transmit and receive beams while transversal filtering on receive enables dynamic focusing. It is shown theoretically that such an implementation is expected to improve the reverberation noise suppression even from reduced-sized reflectors. The simulation results show an improvement in noise suppression up to 9dB for point scatterers in case of using equal transmit and receive beams.