Is there any objective and independent characterization and modeling of soft biological tissues?

The characterization of soft tissue raises several difficulties. Indeed, soft biological tissues usually shrink when dissected from their in vivo location. This shrinkage is characteristic of the release of residual stresses, since soft tissues are indeed often pre-stressed in their physiological configuration. During experimental loading, large extension at very low level of force are expected and assumed to be related to the progressive recruitment and stretching of fibers. However, the first phase of the mechanical test is also aiming at recovering the pre-stressed in vivo behavior. As a consequence, the initial phase, corresponding to the recovering of prestress and/or recruitment of fiberes, is questionable and frequently removed. One of the preferred methods to erase it consists in applying a preforce or prestress to the sample: this allows to easily get rid of the sample retensioning range. However this operation can impact the interpretation of the identified mechanical parameters. This study presents an evaluation of the impact of the data processing on the mechanical properties of a numerically defined material. For this purpose, a finite element simulation was performed to replicate a uniaxial tensile test on a biological soft tissue sample. The influence of different pre-stretches on the mechanical parameters of a second order Yeoh model was investigated. The Yeoh mechanical parameters, or any other strain energy density, depend strongly on any pre- and post-processing choices: they adapt to compensate the error made when choosing an arbitrary level of prestretch or prestress. This observation spreads to any modeling approach used in soft tissues. Mechanical parameters are indeed naturally bound to the choice of the pre-stretch (or pre-stress) through the elongation and the constitutive law. Regardless of the model, it would therefore be pointless to compare mechanical parameters if the conditions for the processing of experimental raw data are not fully documented.