A new detection method for vascular network in medical images

Image detection has been widely used in such fields as geometric measurement, industrial quality inspection, three-dimensional surface detection and so on. Because most of the image detection algorithms are effective for specific problems, there is no kind of detection algorithm can be applied for all image detection problems. An accurate detection of the vascular network in medical images from various organs can help physicians to plan their surgical operations. In this paper, we propose a new adaptive medialness measure method for detection of vascular network. The adaptiveness of the medialness is based on the eigenvectors and eigenvalues of the Hessian matrix of the image. In this method, we describe the measure of medialness and ridge, then we use a toric circular model of a vessel to illustrate our detection, we express the relationship between the size of the structure and its selected scale. From this relationship, we explain the extraction of the local extrema and make a full reconstruction of the vascular network. Eventually synthetic images are used to validate the detailed study and show the behavior of the algorithm under the suggested vascular model.