Purpose: The geometries used to conduct numerical simulations of the biomechanics of the human cornea are reconstructed from images of the physiological configuration of the system, which is not in a stress-free state because of the interaction with the surrounding tissues. If the goal of the simulation is a realistic estimation of the mechanical engagement of the system, it is mandatory to obtain a stress-free configuration to which the external actions can be applied.
Methods: Starting from a unique physiological image, the search of the stress-free configuration must be based on methods of inverse analysis. Inverse analysis assumes the knowledge of one or more geometrical configurations and, chosen a material model, obtains the optimal values of the material parameters that provide the numerical configurations closest to the physiological images. Given the multiplicity of available material models, the solution is not unique.
Results: Three exemplary material models are used in this study to demonstrate that the obtained, non-unique, stress-free configuration is indeed strongly dependent on both material model and on material parameters.
Conclusion: The likeliness of recovering the actual stress-free configuration of the human cornea can be improved by using and comparing two or more imaged configurations of the same cornea.
Material model calibration through indentation test and stochastic inverse analysis
