Force-Sensor-Based Estimation of Needle Tip Deflection in Brachytherapy

Abstract

A virtual sensor is developed for the online estimation of needle tip deflection during permanent interstitial brachytherapy needle insertion. Permanent interstitial brachytherapy is an effective, minimally invasive, and patient friendly cancer treatment procedure. The deflection of the needles used in the procedure, however, undermines the treatment efficiency and, therefore, needs to be minimized. Any feedback control technique to minimize the needle deflection will require feedback of this quantity, which is not easy to provide. The proposed virtual sensor for needle deflection incorporates a force/torque sensor, mounted at the base of the needle that always remains outside the patient. The measured forces/torques are used by a mathematical model, developed based on mechanical needle properties. The resulting estimation of tip deflection in real time during needle insertion is the main contribution of this paper. The proposed approach solely relies on the measured forces and torques without a need for any other invasive/noninvasive sensing devices. A few mechanical models have been introduced previously regarding the way the forces are composed along the needle during insertion; we will compare our model to those approaches in terms of accuracy. In order to conduct experiments to verify the deflection model, a custom-built, 2-DOF robotic system for needle insertion is developed and discussed. This system is a prototype of an intelligent, hand-held surgical assistant tool that incorporates the virtual sensor proposed in this paper.