MODELING OF NEEDLE INSERTION FORCES FOR PERCUTANEOUS THERAPIES

Abstract

This thesis reports on a method for modeling data from needle insertions into soft tissue. Force information from needle insertions was measured and modeled for use in robot-assisted percutaneous therapies. The goal of the work reported here was to develop a new methodology for reality-based data modeling that results in more accurate representation of a needle insertion scenario that could be used as a holistic relative position guide during percutaneous procedures. Insertion force data was collected on bovine livers using the Johns Hopkins University Steady-Hand Robot, and modeled in three parts: force due to capsule stiffness, friction, and cutting. Capsule stiffness is modeled by a nonlinear spring model, friction by a modified Karnopp model, and cutting by the remaining forces, which appear to be constant for a given tissue sample. During robot-assisted procedures, real-time force data can be compared to these models to control puncture of interior structures. Clinical applications of this work include liver ablation, biopsies, and prostate brachytherapy.