In this manuscript, we present a combined experimental and computational technique that can identify the heterogeneous elastic properties of planar soft tissues. By combining inverse membrane analysis, digital image correlation, and bulge inflation tests, we are able to identify a tissue's mechanical properties locally. To show how the proposed method could be implemented, we quantified the heterogeneous material properties of a human ascending thoracic aortic aneurysm (ATAA). The ATAA was inflated at a constant rate using a bulge inflation device until it ruptured. Every 3 kPa images were taken using a stereo digital image correlation system. From the images, the three-dimensional displacement of the sample surface was determined. A deforming NURBS mesh was derived from the displacement data, and the local strains were computed. The wall stresses at each pressure increment were determined using inverse membrane analysis. The local material properties of the ATAA were then identified using the pointwise stress and strain data. To show that it is necessary to consider the heterogeneous distribution of the mechanical properties in the ATAA, three different forward finite element simulations using pointwise, elementwise, and homogeneous material properties were compared. The forward finite element predictions revealed that heterogeneous nature of the ATAA must be accounted for to accurately reproduce the stress-strain response.

Ascending thoracic aortic aneurysms (ATAAs) are focal, asymmetric dilatations of the aortic wall which are prone to rupture. To identify potential rupture locations in advance, it is necessary to consider the inhomogeneity of the ATAA at the millimeter scale. Towards this end, we have developed a combined experimental and computational approach using bulge inflation tests, digital image correlation (DIC), and an inverse membrane approach to characterize the pointwise stress, strain, and hyperelastic properties of the ATAA. Using this approach, the pointwise hyperelastic material properties were identified on 10 human ATAA samples collected from patients undergoing elective surgery to replace their ATAAs with a graft. Our method was able to capture the varying levels of heterogeneity in the ATAA from regional to local. It was shown for the first time that the material properties in the ATAA are unmistakably heterogeneous at length scales between 1 mm and 1 cm, which are length scales where vascular tissue is typically treated as homogeneous. The distributions of the material properties for each patient were also examined to study the inter-and intra-patient variability. Large inter-subject variability was observed in the elastic properties.

A major challenge in the experimental study of aneurysm properties is that the tissues are heterogeneous. When the specimens are not reasonably homogeneous, traditional tests that work under the premise of tissue homogeneity cannot reliably delineate the local conditions at the rupture site. In this work, we investigated the local characteristics of rupture of human ascending aortic aneurysm tissue. We identified the stress, strain, and elastic properties to a submillimeter resolution. Based on the field values, we determined the local conditions - elastic properties, direction of maximum stiffness, stress, strain, energy consumption - at the rupture site. It was found that the tissues consistently cleave in the direction of the maximum stiffness, and generally occurs at the location of highest energy. Since a higher stiffness and higher strain energy indicate a larger recruitment of collagen fibers in the tissue at the location and along the direction of rupture, the work suggests that the recruitment of collagen fibers in the deformation of the tissue is probably essential in aneurysm rupture.

Machine learning was applied to classify tension-strain curves harvested from inflation tests on ascending thoracic aneurysm samples. The curves were classified into rupture and nonrupture groups using prerupture response features. Two groups of features were used as the basis for classification. The first was the constitutive parameters fitted from the tension-strain data, and the second was geometric parameters extracted from the tension-strain curve. Based on the importance scores provided by the machine learning, implications of some features were interrogated. It was found that (1) the value of a constitutive parameter is nearly the same for all members in the rupture group and (2) the strength correlates strongly with a tension in the early phase of response as well as with the end stiffness. The study suggests that the strength, which is not available without rupturing the tissue, may be indirectly inferred from prerupture response features.

No abstract