Implementation and evaluation of hyperelastic model for surgical simulator and navigation

We developed a practical laparoscopic surgical simulator using co-rotated FEM in a linear scheme. This was somewhat of a compromise due to a strong constraint on real-time processing. The spread of surgical simulators and medical simulations for clinical medicine in the near future will impose important demands that cannot be met with this linear scheme. For example, subtle force sensing by forceps used for peeling connective tissues and moving blood vessels is very important for a preoperative surgical simulator, as is precisely predicting the deformation of organs with patient posture during surgery for torocar simulation and surgical navigation. We evaluated several models such as co-rotated FEM, nonlinear FEM, and the hyperelastic model for these advanced real-time medical applications. As a result, we confirmed that the hyperelastic model is the most suitable for the anticipated surgical simulator, and that the co-rotated FEM and nonlinear FEM score almost the same in both processing time and accuracy. In addition, implementation of the hyperelastic model in real time seems possible with current off-the-shelf PCs.