Characterizing the Biomechanical Properties of the Pubovisceralis Muscle Using a Genetic Algorithm and the Finite Element Method.

To better understand the disorders in the pelvic cavity associated with the pelvic floor muscles (PFM) using computational models, it is fundamental to identify the biomechanical properties of these muscles. For this purpose, we implemented an optimization scheme, involving a genetic algorithm (GA) and an inverse finite element analysis (FEA), in order to estimate the material properties of the pubovisceralis muscle (PVM). The datasets of five women were included in this noninvasive analysis. The numerical models of the PVM were built from static axial magnetic resonance (MR) images, and the hyperplastic Mooney-Rivlin constitutive model was used. The material parameters obtained were compared with the ones established through a similar optimization scheme, using Powell's algorithm. To validate the values of the material parameters that characterize the passive behavior of the PVM, the displacements obtained via the numerical models with both methods were compared with dynamic MR images acquired during Valsalva maneuver. The material parameters (c1 and c2) were higher for the GA than for Powell's algorithm, but when comparing the magnitude of the displacements in millimeter of the PVM, there was only a 5% difference, and 4% for the principal logarithmic strain. The GA allowed estimating the in vivo biomechanical properties of the PVM of different subjects, requiring a lower number of simulations when compared to Powell's algorithm.