Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.772859
Title: Clutter noise reduction in ultrafast B-mode image
Author: Mohamed Moubark, Asraf
ISNI:       0000 0004 7960 3137
Awarding Body: University of Leeds
Current Institution: University of Leeds
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Ultrafast ultrasound imaging using plane waves has advanced a myriad of novel ultrasound imaging methods such as ultrafast contrast-enhanced imaging and shear wave elastography. However, due to the lack of transmission focusing and the beam steering operation in plane wave imaging (PWI), the presence of clutter noise on the B-mode image is unavoidable. The reduction of clutter noise is expected to increase the image contrast and spatial resolution, the two main criteria in ultrasound medical B-mode imaging for better diagnosis. Researchers have looked into this problem in depth and proposed many solutions. Many of the proposed solutions come with trade-offs. Attenuating the clutter noise may reduce the frame rate (FR), broaden the main lobe or increase the overall computational complexity. Thus, an advanced solution is warranted, with which reducing the clutter noise will not affect the FR and the image quality can be improved with a low computational complexity. In this thesis, the clutter noise problem has been tackled with various approaches, the first implementing a new filtered delay multiply and sum (FDMAS) beamforming technique, the second being a new compounding method based on autocorrelation and the final being a new version of unsharp masking (UM) filter. In FDMAS, the optimization of the imaging point step in lateral direction has been investigated. By calculating the received echoes with a smaller imaging point in the lateral direction, it helps to lower the side lobe levels and improve the lateral resolution of B-mode images. The proposed compounding method that is based on the autocorrelation process has proved effective to reduce clutter noise even with a very low number of compounding angles. The modified version of the UM filter suites ultrasound B-mode imaging and it provides promising results granting the improved image contrast and resolution without compromising the FR. The benefit of reducing clutter noise on the B-mode image was demonstrated with its application to contour segmentation. Attenuating clutter noise not only speeded up the segmentation process but also benefited the measurement accuracy of the intima media thickness, which has the potential to add diagnostic value for the early detection of stokes, among other vascular diseases.
Supervisor: Freear, Steven Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.772859  DOI: Not available
Share: