Cooperative robotic assistant with drill‐by‐wire end‐effector for spinal fusion surgery

Purpose – The purpose of this paper is to present a surgical robot for spinal fusion and its control framework that provides higher operation accuracy, greater flexibility of robot position control, and improved ergonomics. Design/methodology/approach – A human‐guided robot for the spinal fusion surgery has been developed with a dexterous end‐effector that is capable of high‐speed drilling for cortical layer gimleting and tele‐operated insertion of screws into the vertebrae. The end‐effector is position‐controlled by a five degrees‐of‐freedom robot body that has a kinematically closed structure to withstand strong reaction force occurring in the surgery. The robot also allows the surgeon to control cooperatively the position and orientation of the end‐effector in order to provide maximum flexibility in exploiting his or her expertise. Also incorporated for improved safety is a “drill‐by‐wire” mechanism wherein a screw is tele‐drilled by the surgeon in a mechanically decoupled master/slave system. Finally, a torque‐rendering algorithm that adds synthetic open‐loop high‐frequency components on feedback torque increases the realism of tele‐drilling in the screw‐by‐wire mechanism. Findings – Experimental results indicated that this assistive robot for spinal fusion performs drilling tasks within the static regulation errors less than 0.1  μ m for position control and less than 0.05° for orientation control. The users of the tele‐drilling reported subjectively that they experienced torque feedback similar to that of direct screw insertion. Research limitations/implications – Although the robotic surgery system itself has been developed, integration with surgery planning and tracking systems is ongoing. Thus, the screw insertion accuracy of a whole surgery system with the assistive robot is to be investigated in the near future. Originality/value – The paper arguably pioneers the dexterous end‐effector appropriately designed for spinal fusion, the cooperative robot position‐control algorithm, the screw‐by‐wire mechanism for indirect screw insertion, and the torque‐rendering algorithm for more realistic torque feedback. In particular, the system has the potential of circumventing the screw‐loosening problem, a common defect in the conventional surgeon‐operated or robot‐assisted spinal fusion surgery.