Mechanical characterization of a brain phantom material by combining experiments in the time and frequency domain

Abstract

Modeling the mechanical behavior of brain tissue complements experimental findings about brain diseases and supports the development of predictive tools for diagnosis. The foundation for a reliable continuum-based model is an accurate and comprehensive experimental characterization of the material. Such a characterization is yet limited by inconsistent or contradicting mechanical responses when using different testing techniques. These inconsistencies mainly result from the ultrasoft behavior and biphasic structure of the tissue, which makes it extremely sensitive to changes in the time and length scales. In this study, an approach is presented to overcome the challenge of inconsistent responses and to unify the results from experiments with varying time scales in a continuum-based model. A viscoelastic hydrogel, validated as a brain phantom material, was experimentally characterized over an extended time range. The quasi-static response was investigated at the rheometer with experiments under multiple loading conditions. The behavior in the mid-frequency range was characterized in a vibration analysis at a custom-built vibration table and the response at high frequencies was studied with magnetic resonance elastography. Moreover, the impact of the testing temperature on the mechanical behavior of the hydrogel was analyzed. A hyper-viscoelastic model was calibrated to the time response conducted at the rheometer. As a material model the hyperelastic Ogden model in combination with the time-dependent Prony series was chosen. By addressing the frequency domain with the relaxation times of the Prony series, the frequency-dependent material behavior was included in the modeling approach. To validate this approach, the experimental responses in the mid and high-frequency range were predicted with the calibrated model. The comparison between the modeled and the measured response revealed an excellent prediction of the elastic material behavior, whereas the viscous response may be underpredicted by the model. The results further highlight that the material is very sensitive to temperature changes and therefore temperature should be taken into account in the comparison of different testing techniques.