A Shape Optimization Technique for Medical Images

Abstract

Segmentation and restoration are among the primary objectives in the field of medical image processing. Among many methods available today, derivative method specifically domain based derivative method helps immensely in this field of medical imaging. There was a time-based improvement in the aspects of establishing results in a medical image with the aid of topological derivative. Advantage of this method lies in establishing a link between that of the perturbed domain and that of the original domain and solving for the required metrics in the problem. Treating the original domain as the original image and the diseased image to be the perturbed domain, topological derivative helps to solve the problem. Asymptotic analysis also helps in handling both large and small amount of data based on the big-O notation and small-o notation. It also establishes the supremacy of topological derived image metrics to that of other metrics. Here attempts are made for solving problems in medical images with the help of topological derivative method by adapting minimization procedure. Also, the end results are arrived in an outstanding way with help of the receiver operating characteristic curve.