Direct Estimation of Strains at Carotid Artery Wall Using Autocorrelation of Ultrasound Images

Thumbnail Image

Persistent link to this item

View Statistics

Journal Title

Journal ISSN

Volume Title


Direct Estimation of Strains at Carotid Artery Wall Using Autocorrelation of Ultrasound Images

Published Date




Thesis or Dissertation


Acute ischemic disorders such as myocardial infarction and stroke are the main cause of death in modern world. Therefore, methods to detect them early or evaluate their precursors would be of great importance. Change in arterial stiffness, or rigidity of the wall, is a very important risk factor in cardiovascular diseases. It is believed that stiffness abnormalities start to form before any apparent clinical indication of the disease. Having an efficient, noninvasive tool to measure arterial stiffness would be of great significance in prevention of arterial disorders such as atherosclerosis or determining their severity in patients. Strain imaging is a promising tool for studying the characteristics of living tissue and has been around for over two decades. MR elastography (MRE) and ultrasound elastography are widely used on commercial scanners, but are far from reaching their full potential. The latter has a number of different implementations such as quasi-static, transient, shear wave, and acoustic radiation force impulse (ARFI). Most of these implementations rely on finding the displacement of tissue in 1D or 2D. In quasi-static elastography, spatial gradients of the axial displacement fields are computed to produce “elasticity maps”, which are overlaid on anatomical B-mode images of the organ, e.g. breast. The quality of these maps depend on the displacement estimation method used and the measurement noise that could be amplified by gradient operators. Displacement estimation plays key role in obtaining tissue elastic properties based on pulse-echo ultrasound. Speckle tracking is the most widely used technique for estimating displacement due to coherent nature of ultrasound. Recent advances of 2D speckle tracking have enabled robust 2D displacement estimation with subsample resolution in both axial and lateral dimensions. Despite these advances, the need for filtered gradient operations for strain computation may limit the usefulness of strain mapping of fine tissue structures such as vessel walls. In this research a new method for direct estimation of axial, lateral and shear strains in tissue is developed. This method takes advantage of local autocorrelation function and the relationship between autocorrelation and power spectral density of analytic radiofrequency (RF) signals. The method assumes sufficiently high frame rates to allow modeling of the frame-to-frame tissue displacement as an affine transformation. In 2D analysis, this would be a Jacobian matrix of 4 elements each of them being one strain parameter. The theory is tested in vivo on posterior wall of common carotid artery of a healthy human subject, as well as in vitro for a vessel mimicking phantom. Results indicate better resolution and accuracy in all four strain parameters than correlation and gradient based techniques. Besides strain, which is a determinant of stiffness of artery, there are other parameters that can independently contribute as a risk factor in cardiovascular diseases. In this research, pulse wave velocity (PWV) was measured and shown that, despite relatively small width of imaging plane and relatively low frame rates, PWV could still be derived from strain data with acceptable accuracy. Available pressure data also made it possible to measure wall shear rate (WSR) and wall shear stress (WSS), two indicators of deformation of flow and wall respectively.



University of Minnesota Ph.D. dissertation. November 2015. Major: Electrical Engineering. Advisor: Emad Ebbini. 1 computer file (PDF); x, 112 pages.

Related to




Series/Report Number

Funding information

Isbn identifier

Doi identifier

Previously Published Citation

Suggested citation

Adibi, Yasaman. (2015). Direct Estimation of Strains at Carotid Artery Wall Using Autocorrelation of Ultrasound Images. Retrieved from the University Digital Conservancy,

Content distributed via the University Digital Conservancy may be subject to additional license and use restrictions applied by the depositor. By using these files, users agree to the Terms of Use. Materials in the UDC may contain content that is disturbing and/or harmful. For more information, please see our statement on harmful content in digital repositories.