Blood vessel segmentation for head MRA using branch-based region growing

We propose an algorithm of blood vessel segmentation for MRA data in this paper. Generic region growing, as well as thresholding, is inappropriate for extracting the totality of the vessels on MRA data. This is because of the image properties of the MRA, where the intensity of each pixel in the blood area depends on the amount of blood flow. Moreover, thin vessels are affected by the partial volume effect, which reduces the intensity of vessel parts. Thus, the range of the intensity of the blood vessel in MRA images is not restricted to a small interval but is spread widely. To get correct segmentation results by region growing, the growing condition should be flexibly adapted according to the local characteristics in each ROI. We have designed a branch‐based region growing method for this purpose. Since its growing process is performed on one branch at a time, the growing conditions can be optimized according to the surrounding properties. It is also possible to connect a break point by extending the vessel, which improves segmentation results. By applying this method to five head MRA data sets, the availability of the method has been confirmed. In addition, to evaluate the segmentation result quantitatively, we developed a new evaluation method which utilizes MIP data. © 2005 Wiley Periodicals, Inc. Syst Comp Jpn, 36(5): 80–88, 2005; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/scj.20166