Anticipating the in vivo loading conditions is a challenging aspect of validating the design and durability of cardiovascular implants. Complex implant geometries, multi-axial displacements, device/tissue interactions, and time-dependent factors can make it difficult or impossible to identify a single loading cycle that is representative of the full range of conditions. Computational modeling is often used to reduce a large number of complex conditions into representative alternating and mean strains at locations of maximum strain. A novel approach for estimating the alternating and mean strains during in vivo loading will be presented. The method determines local strains by deforming a finite-element model of an implant into a three-dimensional geometry reconstructed from two-dimensional images of a physically deformed implant. The method is then extended to in vitro fatigue experiments. Two examples are given of a structural heart implant.