DEXTERITY AND AUTONOMY IN MINIMALLY INVASIVE SURGICAL ROBOTICS INTERVENTIONS
Johns Hopkins University
Recent technological advances in Medical Robotics have resulted in the development of a range of new techniques that have reduced patient trauma, shortened hospitalization, and improved diagnostic accuracy and therapeutic outcome. Despite the many appreciated benefits of robot-assisted minimally invasive surgery (MIS) compared to traditional approaches, there are still significant drawbacks associated with these robotic systems including both dexterity and autonomy. The dexterity limitation is associated with the poor accessibility to the areas of interest and insufficient instrument control and ergonomics caused by rigidity of the conventional instruments. In other words, the ability to adequately access different target anatomy is still the main challenge of MIS procedures demanding specialized instrumentation, sensing and control paradigms. Furthermore, to enhance the safety of robot-assisted procedures, current robotics research is also exploring new ways of providing synergistic control between the surgeon and the robot. In this context, the robot can perform certain surgical tasks autonomously under the supervision of the surgeon. However, autonomy requires the robot’s perception and adaptation to dynamically changing environments with the human always in the control loop. One of the main challenges of this problem is the unknown large tissue deformation due to the force generated by tool-tissue interaction making the pre-planning and decision-making very sophisticated. Of note, efforts in automating deformable and unstructured soft tissue surgeries have been limited so far to elemental tasks such as knot tying, needle insertion, and executing predefined motions. To address these challenges, this work covers author’s efforts toward bringing dexterity in robot-assisted minimally invasive surgical procedures, and particularly in orthopedics, using continuum manipulators/soft robots, appropriate sensing units, and control paradigms. To be specific, the author has developed flexible debriding and milling tools that can be integrated with a continuum manipulator. The performance of these tools has been evaluated via extensive experiments. The curved drilling technique using a continuum manipulator as well as design and fabrication of a bendable medical screw are other contributions of the author in bringing dexterity for MIS orthopedic interventions. Furthermore, to control a generic unmodelled continuum manipulator working in obstructed environments, the author has developed two different model-independent data-driven learning and control algorithms. In addition, to overcome the difficulties of various model-based autonomous/semi-autonomous control approaches dealing with deformable tissues, the second part of this research focuses on a priori model-independent data-driven approach to autonomously perform a semi-utonomous/autonomous deformable tissue intervention using the da Vinci Research Kit. The performance of this algorithm has extensively been evaluated in different 2-D/3-D homogeneous and heterogeneous phantoms with the presence of internal and external disturbances. Furthermore, this algorithm has been used for indirect manipulation of an unknown deformable tissue with the goal of semi-autonomous cryoablation of kidney tumors.
Medical robotics, Continuum manipulators, surgical autonomy, orthopedics interventions, minimally invasive surgery, flexible instruments