Modeling and Control of A Flexible Needle in Soft Tissue

In this thesis, methods for improving needle steering during prostate brachytherapy are studied. Brachytherapy is a minimally-invasive, out-patient therapy for prostate cancer treatment with a fast recovery time. Success of brachytherapy depends on precise placement of needle tips in pre-planned locations inside and around the prostate. The main sources of needle steering errors in brachytherapy are needle deflection and tissue deformation. Considering these two factors, the long-term goal of research is to develop fully-automatic robot assisted and image guided strategies for needle steering to reduce targeting errors, which will result in increased success of brachytherapy. Toward this goal, for enhancing the current practice of manual brachytherapy, two approaches are proposed in this thesis. First, using Lagrangian formulation, a novel dynamical model of flexible needle in soft tissue is derived. This model can be used for fully-automatic needle steering; we have studied closed-loop control of the needle in simulations. Second, focusing on semi-automatic needle steering in order to allow for continued involvement of the brachytherapist in the procedure, we develop an experimental set up that closely resembles manual brachytherapy except for an enhancement involving automatic needle deflection adjustment. Simulation and experimental results reveal the merits of the proposed approaches and potentials for future research.