Fast and Automatic Vessel Centerline Detection for MRA

نویسندگان

  • K. Krissian
  • C. F. Westin
  • R. Kikinis
چکیده

The extraction of vessel centerlines allows for better understanding of their topology, and can be used for further registration or quantification. We present a new method to extract the centerlines from Magnetic Resonance Angiographies. We use the gradient and the eigenvalues of the Hessian matrix to interpolate the zero-crossing of the gradient vector in the cross-sectional directions. Then, we connect the points belonging to the same cubes. A second connection step allows for the improvement the global connectivity. Our method is fast, extracts sub-voxel centerlines, and does not rely on a preliminary segmentation stage. Introduction Several methods can be considered in order to obtain the centerlines of vessels. A first class of methods derives a set of voxels that belong to the centerlines and then creates the lines as sets of points located at the center of voxels. These voxels can be obtained by simple thresholding of the image and then applying a skeletonization process, or by a linear multi-scale analysis [3]. A second class of methods tries to extract directly sub-voxel centerlines considering the line as an intersection of two iso-surfaces [2]. Sub-voxel Centerline Extraction Our method belongs to this second class of algorithms. It is based on the extraction of ridges as defined in [l]. Let I be the image as a function from B3 to B, VI be the gradient vector field, H ( I ) be the Hessian matrix with eigenvectors v l , v2, vs, and associated eigenvalues A1 5 XZ 5 X3. The derivatives of the intensity are computed by convolving the image with the derivatives of a Gaussian kernel of standard deviation n. We are interested in the automatic extraction of the points which satisfy the following properties: Improvement of the connectivity Although the algorithm is able to extract most of the centerlines, we post-process the detected lines using connectivity criteria which allows a better continuity of the vessels. We connect two extremities of distinct centerlines M I and Mz, with associated tangents t l and t2 (oriented outward), if and only if the distance between Mi and ,442 is lower than d,, and the two angles (tl , a) and (t2, MzMl) are lower than __f

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

An improved cerebral vessel extraction method for MRA images.

To improve the speed and accuracy of cerebral vessel extraction, a fast and robust method is proposed in this paper. First, volume data are divided into sub-volumes by using octree, and at the same time invalid volume data are eliminated. Second, fuzzy connectedness is introduced to achieve fast cerebral vessel segmentation from 3D MRA Images. The values of gradient and Laplacian transformation...

متن کامل

Automatic correction of gaps in cerebrovascular segmentations extracted from 3D time-of-flight MRA datasets.

OBJECTIVES Exact cerebrovascular segmentations are required for several applications in today's clinical routine. A major drawback of typical automatic segmentation methods is the occurrence of gaps within the segmentation. These gaps are typically located at small vessel structures exhibiting low intensities. Manual correction is very time-consuming and not suitable in clinical practice. This ...

متن کامل

Vessel Centerline Tracking in CTA and MRA Images Using Hough Transform

We proposed a semi-automated method to detect and track vessel centerlines in CTA and MRA images. During the tracking process, we analyze probable bifucation points. The methodology has shown to work well under significant amounts of noise. The main contribution of this work is the use of HT to define the seed point for tracking and detecting the vessel scale without resorting to multiscale ana...

متن کامل

A Deformable Vessel Model with Single Point Initialization for Segmentation, Quantification and Visualization of Blood Vessels in 3D MRA

We deal with image segmentation applied to three-dimensional (3D) analysis of vascular morphology in magnetic resonance angiography (MRA) images. The main goal of our work is to develop a fast and reliable method for stenosis quantification. The first step towards this purpose is the extraction of the vessel axis by an expansible skeleton method. Vessel boundaries are then detected in the plane...

متن کامل

Extracting Vessel Centerlines From Retinal Images Using Topographical Properties and Directional Filters

In this paper we consider the problem of blood vessel segmentation in retinal images. After enhancing the retinal image we use green channel of images for segmentation as it provides better discrimination between vessels and background. We consider the negative of retinal green channel image as a topographical surface and extract ridge points on this surface. The points with this property are l...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2001