RECURSIVE TRACKING OF VASCULAR NETWORKS IN ANGIOGRAMS BASED ON THE DETECTION DELETION SCHEME

A computer algorithm was developed for automated identification of 2-D vascular networks in X-ray angiograms. This was accomplished by using an adaptive tracking algorithm in a three-stage recursive procedure. First, given a starting position and direction, a segment in the vascular network was identified. Second, by filling it with the surrounding background pixel values, the detected segment was deleted from the angiogram. The detection-deletion scheme was employed to prevent the problem of tracking-path reentry in those areas where vessels overlap. Third, all branch points were detected by use of matched filtering along both edges of the vessel. The detected branch points were used as the starting points in the next recursion. The recursive procedure terminated when no new branch point was found. The algorithm showed a good performance when it was applied to angiograms of coronary and radial arteries. To provide a quantitative evaluation, vascular networks identified by the algorithm were compared to those identified by human. The algorithm made some false-negative errors, but very few false-positive errors. The true detection rate for vascular pixels was between 81 and 97% for angiograms with digital subtraction (n = 3) and without subtraction (n = 8), and 75% in a worst case study using low-quality cineangiograms (n = 3). The computational time was on average 3 s on a workstation. The algorithm should be useful for automated detection and quantitative analysis of blood vessel images.