Academic
Publications
Robust Vessel Tree Modeling

Robust Vessel Tree Modeling,10.1007/978-3-540-85988-8_72,M. Akif Gülsün,Hüseyin Tek

Robust Vessel Tree Modeling   (Citations: 9)
BibTex | RIS | RefWorks Download
In this paper, we present a novel method for extracting center axis representations (centerlines) of blood vessels in contrast enhanced (CE)-CTA/MRA, robustly and accurately. This graph-based optimization algorithm which employs multi-scale medialness filters extracts vessel centerlines by computing the minimum-cost paths. Specifically, first, new medialness filters are designed from the assumption of circular/elliptic vessel cross-sections. These filters produce contrast and scale independent responses even the presence of nearby structures. Second, they are incorporated to the minimum-cost path detection algorithm in a novel way for the computational efficiency and accuracy. Third, the full vessel centerline tree is constructed from this optimization technique by assigning a saliency measure for each centerline from their length and radius information. The proposed method is computationally efficient and produces results that are comparable in quality to the ones created by experts. It has been tested on more than 100 coronary artery data set where the full coronary artery trees are extracted in 21 seconds in average on a 3.2GHz PC.
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...In order to examine a CTCA dataset for calcified coronary lesions, the entire vascular tree is segmented with the method presented by Gülsün and Tek [8]...
    • ...Subsequently, a local center-axis description is determined using a minimum-cost path detection [5, 12 ]o n the normalized edge response [8]...

    Matthias Teßmannet al. Automatic detection and quantification of coronary calcium on 3D CT an...

    • ...The costs of all feasible paths can be efficiently computed by using different operators such as the vesselness filters [2], medialness filters [3,4], image Hessian [5], polar intensity...
    • ...It is also difficult to determine the endpoints of the centerlines without producing additional spurious branches, i.e., poor convergence due to leakage towards the non-vascular areas [4]...
    • ...This stage can be considered as a traditional minimal path extraction scheme (described in our previous work [4]) to obtain a coarse estimate of the vessel centerline...
    • ...� ek� =1 , fork =1 , 2, and ε measures the normalized edge response derived from the image gradients, as described in [4]...
    • ...We kindly refer the reader to [4] for an extended discussion and further details on the implementation...
    • ...Due to strong empirical evidence, we anticipate that the proposed approach quantitatively outperforms its predecessor in [4], whose performance was reported in great detail in [1]...

    Hasan Ertan Çetingület al. A Unified Minimal Path Tracking and Topology Characterization Approach...

    • ...A different class of segmentation algorithms uses wavefront propagation or another type of tracking to grow the vessel tree from a number of initial seed points [2, 4-6]...
    • ...Also, experts have visually assessed the correctness of the segmentation [6]...

    Evelien van Dongenet al. Automatic segmentation of pulmonary vasculature in thoracic CT scans w...

    • ...This is also the case with certain tracking methods [7, 8] that rely on threshold based stopping criteria...

    Pechin Loet al. Vessel tree extraction using locally optimal paths

    • ...Recent works proposed multi-hypotheses schemes to alleviate limitations of the local optimization process [1, 2, 3, 4]. The combination of all these factors exacerbates the need for selective and computationally efficient features to evaluate shape and appearance hypotheses on the image...
    • ...Increased robustness comes however at the detriment of accuracy, as discussed in [11, 12]...
    • ...We compare Flux and MFlux to the Hessian-based Vesselness measure from [15] (with α = β =0 .5), the Ribbon measure from [1] and the Core tubular feature from [11]...
    • ...We restrict ourselves to circular cross-sectional areas, where [1] also considered ellipsoids...

    David Lesageet al. Design and Study of Flux-Based Features for 3D Vascular Tracking

Sort by: