Virtual Cutting and Suturing for an Immersive Finite Element Based Bi-Manual Suturing Simulator
Alex Lindblad, University of Washington
Surgical suturing is a skill not taught in a virtual environment due to the inherent computational costs that prohibit realistic behavior to occur in real-time. This research presents new methods for modeling soft-tissue cutting, and medical suturing in an open procedure using optimized methods to update the Schur complement of the constrained system of equations.
Initialization of the cut occurs once the reaction at the tip of the virtual scalpel exceeds the skin’s puncture force, at which point the first set of constraints is released. The cut progresses as the scalpel passes over and releases the other displacement constraints defining the cut. The main difficulty in this procedure is updating the Schur complement as the constraint matrix changes, during release of the constraints. Removal of a constraint is accomplished by zeroing the rows and columns in the Schur complement associated with that constraint, and then placing a “1” on the diagonal of the affected rows.
The needle is modeled as a rigid displacement constraint forcing the tissue to displace according to the needle position. Once the reaction at the tip has exceeded the skin’s puncture force, the needle penetrates into the tissue. At that point the movement is broken into two components, sliding and non-sliding. Non-sliding displacements result in an update of the displacement constraint, while sliding movement results in the needle penetrating further into the tissue. After the needle passes through the tissue, a new displacement constraint is added that changes the relative displacement between the two contact points on the surface of the skin. Updating the Schur complement after the addition of these constraints is done through two vector-matrix products. The goal of this research has been to increase the level of realism in surgical suturing simulators while advancing their capabilities to a point closer to medical integration.
Abstract Author(s): A.J. Lindblad, G.M. Turkiyyah