A Quantitative Evaluation of the Hip Prosthesis Segmentation Quality in X-Ray Images

Abstract

X-ray film images are the main medical diagnosis tool in the evaluation of the fit of the hip prostheses inserted in total hip arthroplasty (THA) procedures. In a computer-aided diagnosis tool, one of the most important operations is the automatic segmentation of the X-ray image into the clinical relevant parts: prosthesis, bone (femur) and soft tissue. The paper investigates the use of several classical adaptive region segmentation techniques, using either the initial pixel luminance space (adaptive histogram thresholding), or an extended feature space (fuzzy C-means) and evaluates the segmentation quality, by the standard detection error and ROC (receiver operating characteristics) curves.