Motion estimation of skeletonized angiographic images using elastic registration

Abstract

An approach for estimating the motion of arteries in digital angiographic image sequences is proposed. Binary skeleton images are registered using an elastic registration algorithm in order to estimate the motion of the corresponding arteries. This algorithm operates recursively on the skeleton images by considering an autoregressive (AR) model of the deformation in conjunction with a dynamic programming (DP) algorithm. The AR model is used at the pixel level and provides a suitable cost function to DP through the innovation process. In addition, a moving average (MA) model for the motion of the entire skeleton is used in combination with the local AR model for improved registration results. The performance of this motion estimation method is demonstrated on simulated and real digital angiographic image sequences. It is shown that motion estimation using elastic registration of skeletons is very successful especially with low contrast and noisy angiographic images.

Digital Topology and Geometry in Medical Imaging: A Survey

“…Skeletonization has been widely used in different medical imaging applications, including arterial motion in cardiac angiographic image sequence [325], [326], feature detection in mammograms [23], [27], vessel segmentation in MR angiography [15], [327], mathematical morphologic image interpolation [328], object characterization [278], stenosis detection [29]- [32], [329]- [331], path finding in colonoscopy [19], [21], [332] and bronchoscopy [18], [20], tree analysis in surgical planning [9] and pulmonary imaging [333], and local structure analysis in trabecular bone imaging [24], [28], [75], [263], [264], [268], [334], [335]. Also, skeletonization has been used to build the correspondence of matching landmarks in a cardiac angiographic image sequence for computation of arterial motion [325], [326].…”

Three-dimensional motion tracking of coronary arteries in biplane cineangiograms

“…The B-solid is the dimension three analog of the onedimensional B-spline curve, and the 2-D B-spline tensor product surface. For a point on the coronary tree at t 0 , the B-solid D t : ℝ 3 → ℝ 3 gives the 3-D displacement vector to that point's position at time t 0 + t. Given a point q = (q x , q y , q z ) ∈ ℝ 3 at time t 0 , its displacement to time t 0 + t is expressed as (12) where the are the n i × n j × n k control points, and the {B. ,p } are the pth-degree Bspline basis functions. The control point density is directly proportional to the amount of local deformation that can be represented by the B-solid.…”

“…For each cardiac phase, a 3-D coronary tree is independently reconstructed. Optical flow [11], binary image elastic registration [12], Kalman snakes [13], and local space search and graph minimization techniques [14] are some of the methods that have been proposed for 2-D vessel tracking. However, tracking the vessels in the projection image space has significant limitations.…”

A dynamic programming solution to tracking and elastically matching left ventricular walls in cardiac cine MRI

“…Three major approaches exist: (i) active contour based methods where an initial contour is refitted near strong local image features by minimizing an energy function [6]; (ii) deformable parametric models with limited degrees of freedom [7,8]; (iii) statistical description and learning based methods of local shape and brightness [9]; and combinations of the above [10,11]. DP optimization has been used in medical imaging; for example in detecting, tracking and matching LV boundaries [12], and matching skeletonized angiographic images [13].…”