Constrained optimal control of needle deflection for semi-manual steering

Brachytherapy is a widely used treatment for patients with localized cancer where high doses of radiation are administered to cancerous tissue by implanting radioactive seeds into the prostate using long beveled-tip needles. Accurate seed placement is an important factor that influences the outcome of the treatment. In this paper, we present and study the suitability of a new framework for computer-assisted needle steering during seed implantation in brachytherapy. The framework is based on a hand held brachytherapy needle steering apparatus. As the needle is manually inserted, the device automatically rotates the needle's base axially at appropriate depths inside tissue in order to control the trajectory of the needle tip towards a target. The steering controller is based on the Rapidly Exploring Random Tree algorithm that updates the necessary manoeuvres online. Experimental results performed in phantom and ex-vivo biological tissues show an average absolute targeting error of 0.56 ±0.62 and 0.48 ±0.48 mm, respectively.