ABSTRACTThe patient-specific biomechanical analysis of the aorta demands the in vivo mechanical properties of individual patients. Current inverse approaches have shown the feasibility of estimating the nonlinear, anisotropic material parameters from in vivo image data using certain optimization schemes. However, since such inverse methods are dependent on iterative nonlinear optimization, these methods are highly computation-intensive, which may take weeks to complete for only a single patient, inhibiting rapid feedback for clinical use. Recently, machine learning (ML) techniques have led to revolutionary breakthroughs in many applications. A potential paradigm-changing solution to the bottleneck associated with patient-specific computational modeling is to incorporate ML algorithms to expedite the procedure of in vivo material parameter identification. In this paper, we developed a ML-based approach to identify the material parameters from three-dimensional aorta geometries obtained at two different blood pressure levels, namely systolic and diastolic geometries. The nonlinear relationship between the two loaded shapes and the constitutive parameters are established by a ML-model, which was trained and tested using finite element (FE) simulation datasets. Cross-validation was used to adjust the ML-model structure on a training/validation dataset. The accuracy of the ML-model was examined using a testing dataset.
Identification of the concentration‐dependent viscoelastic constitutive parameters of gelatin by combining computational mechanics and machine learning
