Numerous models have been developed to describe the mechanical behavior of soft tissues for medical simulation. Verification, validation and evaluation are crucial steps towards the acceptance of simulation results by clinicians. These tasks, often based on comparisons between simulation results and experimental data or other simulations, are difficult because of the wide range of available modeling techniques, the number of possible assumptions, and the difficulty to perform validation experiments. A comparison framework is proposed based on the analysis of the modelisation process and on a generic description of both constitutive elements of a simulation (e.g. geometry, loads, stability criterion) and results (from simulations or experiments). Generic description allows comparisons between different modeling techniques implemented in various simulation platforms. Comparisons can be performed against real experiments, other simulation results or previous versions of a model thanks to the generic description of results and use a set of metrics to quantify both accuracy and computational efficiency. This description also facilitates validation experiments sharing. The usability of the method is shown on several validation and comparison experiments. The framework is then used to investigate the influence of modeling assumptions and parameters in a biomechanical finite element model of an in-vivo tissue aspiration device. This study gives clues towards the improvement of the predictions of the characterization device.
