Theses Doctoral

2-D and 3-D high frame-rate Pulse Wave Imaging for the characterization of focal vascular disease

Apostolakis, Iason Zacharias

Cardiovascular diseases are major causes of morbidity and mortality in Western-style populations. Atherosclerosis and Abdominal Aortic Aneurysms (AAAs) are two prevalent vascular diseases that may progress without symptoms and contribute to acute cardiovascular events such as stroke and AAA rupture, which are consistently among the leading causes of death worldwide. The imaging methods used in the diagnosis of these diseases, have been reported to present several limitations. Given that both are associated with mechanical changes in the arterial wall, imaging of the arterial mechanical properties may improve early disease detection and patient care.
Pulse wave velocity (PWV) refers to the velocity at which arterial waves generated by ventricular ejection travel along the arterial tree. PWV is a surrogate marker of arterial stiffness linked to cardiovascular mortality. The foot-to-foot method that is typically used to calculate PWV suffers from errors of distance measurements and time-delay measurements. Additionally, a single PWV estimate is provided over a relatively long distance, thus inherently lacking the capability to provide regional arterial stiffness measurements. Pulse Wave Imaging (PWI) is a noninvasive, ultrasound-based technique for imaging the propagation of pulse waves along the wall of major arteries and providing a regional PWV value for the imaged artery.
The aim of this work was to enable PWI to provide more localized PWV and stiffness measurements within the imaged arterial segment and to further extend it into a 2-D and 3-D technique for the detection and monitoring of focal vascular disease at high temporal and spatial resolution. The improved modality was integrated with blood flow imaging modalities aiming to render PWI a comprehensive methodology for the study of arterial biomechanics in vivo.
Spatial information was increased with the introduction of piecewise PWI. This novel technique was used to measure PWV within small sub-regions of the imaged vessel in murine aneurysmal (n = 8) and atherosclerotic aortas (n = 11) in vivo. It provided PWV and stiffness maps while capturing the progressive arterial stiffening caused by atherosclerosis. PWI was further augmented with a sophisticated adaptive algorithm, enabling it to optimally partition the imaged artery into relatively homogeneous segments, automatically isolating arterial stiffness inhomogeneities. Adaptive PWI was validated in silicone phantoms consisting of segments of varying stiffness and then tested in murine aortas in vivo.
Subsequently, the conventional tradeoff between spatial and temporal resolution was addressed with a plane wave compounding implementation of PWI, allowing the acquisition of full field of view frames at over 2000 Hz. A GPU-accelerated PWI post-processing framework was developed for the processing of the big bulk of generated data. The parameters of coherent compounding were optimized in vivo. The optimized sequences were then used in the clinic to assess the mechanical properties of atherosclerotic carotids (n=10) and carotids of patients after endarterectomy (n=7), a procedure to remove the plaque and restore blood flow to the brain. In the case of atherosclerotic patients undergoing carotid endarterectomy, the results were compared against the histology of the excised plaques. Investigation of the mechanical properties of plaques was also conducted for the first time with a high-frequency transducer (18.5 MHz).
Additionally, 4-D PWI was introduced, utilizing high frame rate 3-D plane wave acquisitions with a 2-D matrix array transducer (16x16 elements, 2.5 MHz). A novel methodology for PWV estimation along the direction of pulse wave propagation was implemented and validated in silicone phantoms. 4-D PWI provided comprehensive views of the pulse wave propagation in a plaque phantom and the results were compared against conventional PWI. Finally, its feasibility was tested in the carotid arteries of healthy human subjects (n=6). PWVs derived in 3-D were within the physiological range and showed good agreement with the results of conventional PWI.
Finally, PWI was integrated with flow imaging modalities (Color and Vector Doppler). Thus, full field-of-view, high frame-rate, simultaneous and co-localized imaging of the arterial wall dynamics and color flow as well as 2-D vector flow was implemented. The feasibility of both techniques was tested in healthy subjects (n=6) in vivo. The relationship between the timings of the flow and wall velocities was investigated at multiple locations of the imaged artery. Vector flow velocities were found to be aligned with the vessel’s centerline during peak systole in the common carotid artery and interesting flow patterns were revealed in the case of the carotid bifurcation
Consequently, with the aforementioned improvements and the inclusion of 3-D imaging, PWI is expected to provide comprehensive information on the mechanical properties of pathological arteries, providing clinicians with a powerful tool for the early detection of vascular abnormalities undetectable on the B-mode, while also enabling the monitoring of fully developed vascular pathology and of the recovery of post-operated vessels.


  • thumnail for Apostolakis_columbia_0054D_14750.pdf Apostolakis_columbia_0054D_14750.pdf application/pdf 12.4 MB Download File

More About This Work

Academic Units
Biomedical Engineering
Thesis Advisors
Konofagou, Elisa E.
Ph.D., Columbia University
Published Here
July 1, 2018