Online identification of abdominal tissues in vivo for tissue-aware and injury-avoiding surgical robots

Abstract

This work presents a “smart” robotic surgical grasper capable of identifying tissue during the early stages of a grasp, allowing automated prevention of grasper-induced tissue crush injuries. It employs no additional sensors beyond signals already present in surgical robots. An estimation algorithm using an extended Kalman filter (EKF) is employed for a nonlinear tissue dynamic model, which is investigated in silico as well as in vivo and in situ on porcine models. Results show that while the approach is sensitive to initial conditions, tissue can be identified during the early stage of a typical grasp.