Vessel Surface Topology Extraction from Noisy White and Black Blood TOF Angiography Volumes using Scale-space Framework

J.S. Suri (USA), K. Liu (PRC), L. Kasuboski (USA), S. Singh (UK), and S. Laxminarayan (USA)

Keywords

Magnetic Resonance Angiography (MRA), Black Blood Angiography (BBA), Prefiltering, Median, Directional, ScaleSpace, Ellipsoidal, Vessels, Arteries, Veins, Quan tification, Stenosis, MIP

Abstract

Segmentation and surface display of the blood vessels in noisy white and black blood angiography is a tough prob lem. (see Suri et al. [1], Suri et al. [2] and [3]). This paper presents a scale-space framework along with divide-and conquer approach for vessel surface topology extraction. This algorithm consists of two stages: In stage one, a 3-D non-vascular tissues are removed using scale-space frame work using Suri’s algorithm (see Suri et al. [4]). Here, raw MR angiographic volume is first converted to isotropic volume followed by 3-D higher order separable Gaussian derivative convolution with known scales to generate edge volume. The edge volume is then run by the directional processor at each voxel where the eigenvalues of the 3-D ellipsoid are computed. The vessel score per voxel is then estimated based on these three eigenvalues which suppress the non-vasculature and background structures yielding the filtered volume. The second stage consists constant density surface extraction using a divide-and-conquer approach for developing the connectivity of the scale-space filtered vol ume. The stage-I output is ray-cast to generate the maxi mum intensity projection images for display. The stage-II output is displayed by surface rendering using Geomview package. The system is run over 20 patient studies from different parts of the body such as: brain, abdomen, kidney and knee/ankle. The computer program takes around 150 seconds of processing time per study for a study for a data size of ½¾¢ ½¾¢½ which includes the complete perfor mance evaluation. We also compare our strategy with the recently published MR filtering algorithms by Alexander et al. ’s [8] and Sun et al. ’s [7].

Important Links:



Go Back