A Computationally Efficient Approach for Estimation of Tissue Material Parameters from Clinical Imaging Data Using a Level Set Method

This study proposes a computational method for estimating in vivo mechanical properties of tissues using clinical imaging data. In particular, a new level-set-based objective functional to compare a target and estimated shape of a tissue structure is introduced, along with its integration into an optimization-based approach for inverse material parameter estimation. The approach employs a continuous shape comparison metric using signed distance functions and combines the adjoint method for efficient gradient-based optimization. Simulated inverse problems based upon estimating cardiac ventricular wall stiffness from untagged imaging and hemodynamic data are used to assess the capability of the proposed approach. The results show that the proposed method is able to consistently and effectively minimize the shape-based objective functional to estimate material parameters. The minimization of this shape difference is capable of providing relatively accurate estimates of material parameters, although naturally depending on the sensitivity of the shape change to the particular parameters, and the process is tolerant to the inclusion of model error. Thus, the approach has the potential capability to provide estimates of in vivo mechanical properties of tissues from the shape of the tissue structure as can be directly estimated from imaging data.