Development of a general purpose cooperative shared-control robot assistant for performance of surgical procedures involving bone machining

Abstract

Currently, the interest on robotics for healthcare applications has grown rapidly, due to the remarkable advantages that robotic technologies introduce into the medical field such as enhancing the precision and accuracy of the medical practitioner, providing remote healthcare services with audiovisual and tactile feedback, quantitative assessment of performance, realization of pre-operative planning of interventions and automation of repetitive tasks. Among the various fields of medicine in which robotic devices can be of use, the denomination robotic surgery is used to refer to those technological developments including robotic devices in the performance of surgical procedures. Those have the potential to improve the outcome of surgical interventions by ensuring levels of precision and dexterity unattainable by other means, thus resulting in safer procedures for the patients (reduced risk of infection, less blood loss and scarring due to performance of minimally invasive interventions). This research work presents an innovative surgical workstation developed at CEIT in collaboration with CUN that is intended for general surgical procedures involving bone machining. The performance of the platform has been optimized for a surgical procedure that can be greatly benefitted from the introduction of a surgical robot assistant, which is transpedicular spinal fusion. The key feature of this platform is that it can work under a number of complementary operation paradigms. For instance, it can work in combination with a customized surgical planner and navigation system, which transmit to the robot the optimal approach for performing the surgical task. However, the main operating mode is the cooperative shared-control paradigm, in which the surgeon physically holds the robot assistant (hands-on robot) and the exerted forces (measured by a force/torque sensor) are proportionally translated to motions of the robot. A distinctive characteristic of cooperative devices (which is shared with teleoperated robots) is that trajectories are generated in real-time. Furthermore, for the case of robotic surgery robots usually work in a highly dynamic, unstructured environment that is not characterized a priori so global information is not usually available for optimization of control algorithms. This work is concerned with the development and improvement of the robot assistant integrated in the surgical system, by designing and building the necessary hardware elements as well as conceiving and programming suitable control algorithms according to the operation paradigms implemented in the system. In this sense, ensuring a stable and robust performance of the system is vital for validating the surgical platform. Therefore, several control algorithms are proposed in this work, which have been conceived for operation under the online trajectory generation paradigm, that allow a safer and more stable operation of the robot assistant: a strategy for management of unstable singular configurations of the manipulator and resolution of the inverse kinematics problem is introduced in this work; a suitable adaptive admittance controller for working under the cooperative operation paradigm that allows tremor filtering, correction of the intentional trajectory error and adjustment to human operator force and motion characteristics is presented too; finally, a virtual fixtures module, developed for guidance during surgical procedures is implemented for safer and accurate performance of surgery. Also, an automated rotary tool for drilling operations designed for this surgical platform is proposed in this work. Given the possible incidence of thermal osteo-necrosis phenomena during the bone machining procedures, a predictive algorithm for assessment of these phenomena has been developed in order to control the performance of the automated drilling tool