ULTRASOUND SPECKLE TRACKING STRAIN ESTIMATION IN CAROTID ARTERY PLAQUE PHANTOM WITH SONOMICROMETRY VALIDATION

• 1.  IntroductionCarotid artery plaque characterization is critical for the prevention of ischemic events. Since plaque stiffness has shown to correlate with plaque vulnerability, quantification of plaque strain throughout the heart cycle would be a useful diagnostic tool. Our previous work encompassed the development and validation of a 2D speckle tracking (ST) algorithm to evaluate arterial stiffness by measuring strain in the carotid artery wall in silico, in vitro, and in vivo. The focus of previous studies has been to quantify plaque strain in the radial direction but lack validation against a ground truth measurement. Our objective was to validate radial and longitudinal strain in plaques via sonomicrometry (sono), and compare the measured plaque and arterial wall strain. 2.  MethodThree carotid artery phantoms with soft wall inclusions, mimicking a vulnerable plaque, were constructed (10% polyvinyl alcohol (PVA), 3% graphite) by exposing the vessel and plaque to three and one freeze-thaw cycles (12h freeze, 12h thaw) respectively, see Fig. 1a. The phantoms were embedded in a tissue mimicking mixture (3% Agar, 4% graphite) at approximately 1cm depth with a pump (CompuFlow 1000 MR) connected to the phantom lumen simulating the carotid blood flow. B-mode cineloops (GE Vivid E9, 9LD linear transducer, 10 MHz, 42 fps) recorded the vessel movement at 20 and 30 mL/s peak flows. The radial and longitudinal deformation of the plaque and vessel wall was estimated by an in house 2D ST (kernel size 5x2 wavelengths) algorithm throughout two consecutive cycles. The region of interest was adjusted according to the plaque size. Sono crystals were placed on the plaque and vessel wall and used as a reference of truth. 3.  ResultsFig. 1b and 1c show sample radial and longitudinal strain curves of a phantom with 20mL/s lumen flow with good agreement between sono and ST. A strong correlation was found at radial (r=0.67, p=0.03) and longitudinal peak systolic strain (r=0.84, p<0.001) between sono and ST. The plaque exhibited 47,3% (SD 27,4%) greater radial and 62,3% (SD 83,5%) longitudinal peak strain than the arterial wall when measured with ST. These preliminary data show that it is possible to measure radial and longitudinal strain in plaques; however, more extensive analysis is required as is the feasibility in vivo. 

Ämnesord

TEKNIK OCH TEKNOLOGIER  -- Medicinteknik -- Medicinsk bildbehandling (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Medical Engineering -- Medical Image Processing (hsv//eng)

Publikations- och innehållstyp

ref (ämneskategori)

kon (ämneskategori)