A method of lumbar pedicle screw placement optimization applied to guidance techniques

Abstract Pedicle screws are an established method for spinal segmental fixation. To pursue high accuracy and minimally invasive, various different guidance techniques have developed. However, the preoperative screw position plans are determined by manual selection which demands complex operations and costs a lot of time. In addition, the current paths planning only consider the position, without taking the postoperative screws retention into consideration. In order to solve the problems, a new method was proposed to plan the lumbar pedicle screw placement automatically. Firstly, identify pedicle area and establish initial path based on the current vertebrae segmentation technology. After that, optimize path via the improved boundary limited Nelder–Mead simplex algorithm, and get the highest firmness path in theory. We considered that both the accuracy of position and the firmness of postoperative screws, achieved a fully automatic optimal path planning, which can effectively assist operators, improve the firmness of screw placement.