Robot control by fluoroscopic guidance for minimally invasive spine procedures

This paper describes an algorithm for automatic needle placement using a six degree-of-freedom (DOF) robot during minimally invasive spine procedures. This automatic needle placement is achieved by a closed-loop position control of the robot end effector provided by computer vision. The term fluoroscopy servoing is given to this process since an X-ray fluoroscope is used as the visualization tool. A software application was developed to segment the needle in the X-ray images to find its position, which is then used as feedback to control the robot and move it to a desired target point. A cadaver study was performed to test the accuracy of this application. Twenty trials were performed and the robot reached the desired target point with an average distance error of 1.21 mm.