A vessel segmentation method for serialized cerebralvascular DSA images based on spatial feature point set of rotating coordinate system.

Cerebrovascular pathology is one of the main fatal diseases which seriously affect the human's health. Extracting the accurate image of cerebral vascular tissue is the key of clinical diagnosis. However, the motion artifacts in DSA images seriously affected the quality of vascular subtraction image. In this paper, an automatic and accurate segmentation method is presented to extract the vascular region in the live image of brain. Firstly, a coarse registration for the live image and the mask image is implemented. And then, the SIFT algorithm is utilized to detect geometrical feature points in the serialized subtraction images. After that, a spatial model of rotating coordinate system and a calculative strategy of contextual information are designed to eliminate the error feature points. Finally, based on a dynamic threshold method, the blood vessel image can be obtained by region growing. The context information in the adjacent subtraction images is fully used. The experimental result shows that the segmented cerebral vascular image is satisfactory. This method can provide accurate vessel image data for the clinical operation based on DSA interventional therapy.