Optimization of the 4D-CECT postprocessing settings for accurate strain mapping of vascular tissues

Details

Supervisors

Kerckhofs, Greet ; Marco Castellaro

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] : ingénieur civil biomédical, à finalité spécialisée

Abstract

Cardiovascular pathologies represent all around the word an incumbent problem causing every year the death of millions of people. The perfect functioning of the heart and blood vessels is fundamental for the healthy maintenance of the body. The mechanical behaviour of the vessels’ tissue is strongly linked to their microstructure, but this can be harmed by the damaging effects of pathologies. More and more interest raised through the years about the cardiovascular anatomy in order to deepen the relationship occurring between microstructure of the tissues and cardiovascular diseases. The progress of the technology led to the development of many imaging tools. Microfocus X-rays computed tomography (microCT) is a non-destructive approach able to resolve micrometre scale structure. The combination of this technology with in situ mechanical tests (i.e., 4DmicroCT) allows to describe the mechanical response of the material analysed when loaded or stressed. In material science this field is widely covered, but in biomedical field there still necessity of explore it since many difficulties are still not overcome. Hard and mineralized tissues lend themself well to 4DmicroCT analysis; on the contrary soft tissue such as arteries’ tissue, because of their low X-rays attenuation makes this approach more complex. To overcome this issue, the introduction of contrast agents has been implemented (referred to as contrast-enhanced microCT or CECT). This now allows to improve the acquisition of 3D images and appreciate the microstructure also of soft tissues while the structure in analysis is deformed (i.e., 4D-CECT). 4D-CECT represent a valid tool to investigate the microstructure of the blood vessels and of their biomechanics, driving to the achievement of the principal goal: clarify the relationship between microstructure and pathology effects The further combination of 4DmicroCT with specific techniques, such as the Digital Image Correlation (DIC) or the Digital Volume Correlation (DVC), bring to the possibility to quantify the displacement and the strain experienced. Despite so, in literature still there is the necessity of investigate this field in particular to well define the post-processing imaging procedure and the best way to tune the image acquisition parameters in order to obtain the best image quality possible, but without damaging the tissue and alter its mechanical properties because of the X-rays radiation dose and the loading process. This study aims to explore the postprocessing analysis, focusing on the DVC technique. Two different types of datasets, synthetic sheets originated via Python script aiming to mimic the elastin fibres and porcine aorta specimen microCT scans, are virtually deformed and compute the displacement field. The first main goal is to understand if the synthetic sheets could be a reliable tool able to faithfully reproduce the behaviour of the synthetic sheets. The second main goal is to understand how to improve the DVC, analysing the different factors that could influence it and how the different acquisition parameters could affect the image quality. Was possible to assess that the virtual deformation, both on synthetic and real datasets, coupled with the DVC analysis work properly. On the contrary, more difficult was working with the correlation of two distinct volumes generated with Python script. This master thesis work could represent a good starting point for future improvement of this topic, to get acquaintance with the use of the synthetic sheets as tool for the enhancing of the DVC technique.