A Probabilistic and Anisotropic Failure Metric for Ascending Thoracic Aortic Aneurysm Risk Stratification

Abstract

Experimental studies have shown that aortic wall tensile strengths in circumferential and longitudinal directions are different (i.e., anisotropic), and vary significantly among patients with aortic aneurysm. To assess aneurysm rupture and dissection risk, material failure metric of the aortic wall needs to be accurately defined and determined. Previously such risk assessment methods have largely relied on deterministic or isotropic failure metric. In this study, we develop a novel probabilistic and anisotropic failure metric for risk stratification of ascending thoracic aortic aneurysm (ATAA). To this end, uniaxial tensile tests were performed using aortic tissue samples of 84 ATAA patients, from which a joint probability distribution of the anisotropic wall strengths was obtained. Next, the anisotropic failure probability (FP) based on the Tsai-Hill (TH) failure criterion was derived. The novel FP metric, which incorporates uncertainty in the anisotropic failure properties, can be evaluated after the aortic wall stresses are computed from patient-specific biomechanical analysis. For method validation, "ground-truth" risks of additional 41 ATAA patients were numerically-reconstructed using corresponding CT images and tissue testing data. Performance of different risk stratification methods (e.g., with and without patient-specific hyperelastic properties) was compared using p-value and receiver operating characteristic (ROC) curve. The results show that: (1) the probabilistic FP metric outperforms the deterministic TH metric; and (2) patient-specific hyperelastic properties can help to improve the performance of probabilistic FP metric in ATAA risk stratification.