Characterization of robotic needle insertion and rotation in artificial and ex vivo tissues

Abstract

Accurate needle insertion to reach targets in deformable tissue relies on an understanding of tissue and needle-tissue interaction mechanics. Toward the development of clinically relevant robotic needle insertion, this work addresses (1) the differences and similarities between needle insertion into artificial and ex vivo tissues, and (2) the use of needle rotation data to develop needle-tissue interaction models that can be used to predict the mechanics of needle insertion. Due to challenges associated with working with living tissue, artificial and ex vivo tissues are often employed when modeling and testing robotic needle insertion. Force and motion data recorded during robotic needle insertion into three artificial tissues (PVC rubber, porcine gelatin, and Gelzan) and three ex vivo tissues (chicken breast, calf liver, and pig liver) demonstrated that the mechanics of needle insertion into artificial tissues are distinct from those of ex vivo tissues, due to differences in friction, stiffness, and relaxation properties. Data from needle rotation was used to fit a friction model that describes needle insertion; this approach could enable acquisition of needle-tissue interaction models for planning and control as a minimally invasive first step in a robot-assisted needle insertion procedure.