Two-dimensional speckle tracking using zero phase crossing with Riesz transform

Mohamed Almekkawy, Yasaman Adibi, Fei Zheng, Emad Ebbini, Mohan Chirala

Research output: Contribution to journalConference articlepeer-review

14 Scopus citations

Abstract

Ultrasound speckle tracking (ST) provides robust estimates of fine tissue displacements along the beam direction due to the analytic nature of echo data. A multi-dimensional ST method (MDST) with subsample accuracy in all dimensions is introduced. The algorithm is based on the gradient of the magnitude and the zero-phase crossing of 2D complex correlation of the generalized analytic signal. The generalization method utilizes the Riesz transform, which is the vector extension of the Hilbert transform. Robustness of the tracking algorithm is investigated using realistic synthetic data sequences created with (Field II) for which the bench mark displacement was known. In addition, the new MDST method is used in the estimation of the flow and surrounding tissue motion on human carotid artery in vivo. The data were collected using a linear array probe of a Sonix RP ultrasound scanner at 325 frames per second (fps). The vessel diameter was calculated from the upper and lower vessel wall displacements, and clearly showed a blood pressure wave-like pattern. The results obtained show that using the Riesz transform produces a more robust estimation of the true displacement of the simulated model compared to previously published results. This could have significant impact on strain calculations near vessel walls.

Original languageEnglish (US)
Article number020004
JournalProceedings of Meetings on Acoustics
Volume22
Issue number1
DOIs
StatePublished - Jan 1 2015
Event168th Meeting of the Acoustical Society of America - Indianapolis, United States
Duration: Oct 27 2014Oct 31 2014

Bibliographical note

Publisher Copyright:
© 2015 Acoustical Society of America.

Fingerprint

Dive into the research topics of 'Two-dimensional speckle tracking using zero phase crossing with Riesz transform'. Together they form a unique fingerprint.

Cite this