Efficient collision detection for soft tissue simulation in a surgical planning system

In the field of cranio-maxillofacial surgery, there is a huge demand from surgeons to be able to automatically predict the post-operative face appearance in terms of a pre-specified bone-remodeling plan. Collision detection is a promising means to achieve this simulation. In this paper, therefore, an efficient collision detection method based on a new 3D signed distance field algorithm is proposed to accurately detect the contact positions and compute the penetration depth with the moving of the bones in the simulation, and thus the contact force between the bones and the soft tissues can be estimated using penalty methods. Thereafter, a nonlinear finite element model is employed to compute the deformation of the soft tissue model. The performance of the proposed collision detection algorithm has been improved in memory requirements and computational efficiency against the conventional methods. In addition, the proposed approach has the superior convergence characteristics against other methods. Therefore, the usage of the collision detection method can effectively assist surgeons in automatically predicting the pos-operative face outline.