Extraction of line properties based on direction fields
- Siemens AG, Munich (Germany). Corporate Research and Development
The authors present a new set of algorithms for segmenting lines, mainly blood vessels in X-ray images, and extracting properties such as their intensities, diameters, and center lines. The authors developed a tracking algorithm that checks rules taking the properties of vessels into account. The tools even detect veins, arteries, or catheters of two pixels in diameter and with poor contrast. Compared with other algorithms, such as the Canny line detector or anisotropic diffusion, the authors extract a smoother and connected vessel tree without artifacts in the image background. As the tools depend on common intermediate results, they are very fast when used together. The results will support the 3-D reconstruction of the vessel tree from stereoscopic projections. Moreover, the authors make use of the line intensity measure for enhancing and improving the visibility of vessels in 2-D X-ray images. The processed images are intended to support radiologists in diagnosis, radiation therapy planning, and surgical planning. Radiologists verified the improved quality of the processed images and the enhanced visibility of relevant details, particularly fine blood vessels.
- OSTI ID:
- 207908
- Journal Information:
- IEEE Transactions on Medical Imaging, Vol. 15, Issue 1; Other Information: PBD: Feb 1996
- Country of Publication:
- United States
- Language:
- English
Similar Records
WE-DE-207A-01: Parallels in the Evolution of X-Ray Angiographic Systems and Devices Used for Minimally Invasive Endovascular Therapy
WE-DE-207A-02: Advances in Cone Beam CT Anatomical and Functional Imaging in Angio-Suite to Enable One-Stop-Shop Stroke Imaging Workflow