Dissertation
Rupture risk assessment for ascending thoracic aortic aneurysms: macroscopic rupture pattern and microstructural connection
University of Iowa
Doctor of Philosophy (PhD), University of Iowa
Summer 2022
DOI: 10.25820/etd.006487
Abstract
Ascending thoracic aortic aneurysm (ATAA), a local dilatation in the thoracic aorta, carries one of the highest mortality rates among vascular diseases when it ruptures. Currently, the guideline for determining whether surgical intervention should be performed is given by the maximum-diameter criterion. However, this simple global criterion is controversial since it mayunderestimate the rupture risks of small aneurysms. A reliable evaluation of the rupture risk for ATAA is needed.
Aneurysm rupture is a complicated processes involving events that occur at different length scales, the
micro-level mechanisms related to the rupture remains unclear. At the tissue scale, currently, there is no easy way to obtain patient-specific tissue strength. To fill these gaps, the overall goal of our study is to explore new approaches for rupture risk assessment of ATAA.
In the first part of this work, machine learning (ML) was employed to identify patterns of macroscopic response associated with rupture. The study was carried out using bi-axial tension-strain data collected from inflation test over ATAA specimens. The stress and strain distributions were determined via a combined experimental and computational method. As a first step, ML models were employed to predict the tissue local strength using features in the low strain phase of the response curve. This is a supervised learning approach, the measured tissue strength and response features derived from the early phase of stress-strain curves at the rupture site were fed to train the ML models. The results indicate that the local strength of ATAA tissue could be reliably predicted from the response features, suggesting that the response features could be exploited for assessing the rupture risk of ATAA. On this basis, a data-driven approach was developed to differentiate the response patterns at the rupture and nonrupture site. It was hypothesized that the region where rupture is prone to initiate is associated with a {“high-level of stress build-up”}. To enable the application of our approach in {\it in vivo} settings, only response data in the physiological pressure range were used. Results suggest that the ATAA peak rupture risk regions (PRR) can be reasonably captured, implying that the {\it “stress build-up level”} can be a macroscopic pattern for estimating the rupture risk.
The second part of the work aimed at exploring the associations of micro-structural properties with macroscopic rupture patterns identified previously. It is acknowledged that the progressive recruitment of wavy collagen fibers is a prominent feature in the early phase of response. We hypothesized that the macroscopic patterns discovered earlier are associated with alternated recruitment response. To test the hypothesis, a kinematic average framework for considering the collagen recruitment in hyperelastic constitutive modeling was proposed. An effective stretch, which is a kinematic variable representing the true stretch of the crimped collagen fibers at the tissue-scale, was introduced. Hyperelastic models were formulated using strain invariants derived from the effective stretch. Within the proposed framework, a family of constitutive models accounting for micro-structural properties, including the collagen undulation and separated elastic properties for elastin and collagen, were proposed and applied to describe the uniaxial and biaxial responses of ATAA. The models demonstrated excellent descriptive performance in both data groups. Results from these models indicate that the use of effective stretch alone can adequately capture the effect of the fiber recruitment without invoking phenomenologicalparameters. The rupture properties were investigated in light of the new models. Results suggest that local regions with straighter fibers and/or low collagen density/stiffness tend to rupture earlier due to quicker fiber recruitment. Most importantly, it was found that the ATAA samples ruptured at roughly the same strain level after factoring out the uncrimping stretch. This finding inspired the construction of a strain-based multiscale rupture criterion which takes into account the collagen waviness properties at the micro-level and the macroscopic tissue deformation for ATAA.
Details
- Title: Subtitle
- Rupture risk assessment for ascending thoracic aortic aneurysms: macroscopic rupture pattern and microstructural connection
- Creators
- Xuehuan He
- Contributors
- Jia Lu (Advisor)Stephane Avril (Committee Member)Sharif Rahman (Committee Member)Madhavan L Raghavan (Committee Member)Hiroyuki Sugiyama (Committee Member)
- Resource Type
- Dissertation
- Degree Awarded
- Doctor of Philosophy (PhD), University of Iowa
- Degree in
- Mechanical Engineering
- Date degree season
- Summer 2022
- Publisher
- University of Iowa
- DOI
- 10.25820/etd.006487
- Number of pages
- xviii, 237 pages
- Copyright
- Copyright 2022 Xuehuan He
- Comment
This thesis has been optimized for improved web viewing. If you require the original version, contact the University Archives at the University of Iowa: https://www.lib.uiowa.edu/sc/contact/.
- Language
- English
- Description illustrations
- illustrations (some color)
- Description bibliographic
- Includes bibliographical references (pages 216-237).
- Public Abstract (ETD)
Ascending thoracic aortic aneurysm (ATAA) is a serious health risk since it tends to rupture, causing life-threatening internal bleeding. Currently, clinical decision about whether to perform surgically intervention is based on the aneurysm diameter. However, the diameter criterion is known to be controversial. The goal of this work is to explore alternative approaches of assessing aneurysm rupture risk and also to understand the mechanisms underlying aneurysm rupture. In the first part of the work, machine learning was employed to identify patterns of macroscopic response associated with rupture. The work showed that the local strength and rupture risk of ATAA tissue could be reliably predicted from response features. The second part of the work focused on exploring the micro-structural connection of aneurysm rupture. Constitutive models which could account for micro-structural properties, including the collagen waviness and separated elastic properties for elastin and collagen, were firstly proposed and applied to characterize the responses of ATAA tissue. The rupture properties were investigated based on these models. The rupture appears to initiate at the local region with straighter fibers and/or lower collagen density due to quicker fiber recruitment. More importantly, it was found that, when the uncrimping stretch of wavy collagen fibers was separated out, the ATAA samples ruptured at roughly the same strain level. This finding inspired the development of a multi-scale rupture metric which takes into account fiber waviness properties and macroscopic tissue deformation, for assessing the rupture risk of ATAA.
- Academic Unit
- Mechanical Engineering
- Record Identifier
- 9984285153802771
Metrics
2 File views/ downloads
47 Record Views