Design and Implementation of Gradient Vector Flow Snake to Detect a Reference Object in Pelvic X-Rays for Preoperative Total Hip Arthroplasty Planning Application

摘要

The main interest of this research project is to promote automation in performing preoperative planning for hip joint replacement surgery using a special medical image viewing software, ViewPro™. Preoperative planning is performed to carefully prepare the surgery and to accurately select the hip implants. The first step of preoperative planning is to calibrate the x-ray image to adjust the magnification factor. A femoral head implant is used as magnification factor reference. Automation is introduced by developing an algorithm to automatically detect this reference object in the image. The algorithm used for performing the automatic detection of the reference object is gradient vector flow (GVF) snake. A study has been performed to compare the newly developed semiautomatic algorithm to the old manual calibration algorithm. The results show a close relation between the two algorithms (less than 1% of average relative difference). It is concluded that the developed semiautomated algorithm can be used as an alternative for performing the manual calibration.