Segmentation of the aorta from CT and MR data is important in order to quantitatively assess diseases of the aorta including aortic dissection and distention of aortic aneurysm, among others. In this paper, we propose a segmentation method to extract exact the 3D boundary of the aorta via graph-cuts segmentation. The graph-cuts technique is able to avoid local minima with global optimization and can be applied to 3D and higher dimension with fast computation. We performed 3D segmentation using this method for five CT data sets. The user selects seed points for aorta region as 'object' and surrounding tissues as 'background' on an axial slice of the 3D CT data and the algorithm calculates the cost of n-link (neighborhood-link) and t-link (terminal-link), and computes the minimum cut separating the aorta from the background by applying the max-flow/min-cut algorithm. Results were validated against manually traced aorta boundaries. The mean Dice Similarity Coefficient for the five 3D segmentations was 0.9381. The 3D segmentation took less than five minutes for data sets of size 512×512×244 to 512×512×284.
Lung boundary image segmentation is important for many tasks including for example in development of radiation treatment plans for subjects with thoracic malignancies. In this paper, we describe a method and parameter settings for accurate 3D lung boundary segmentation based on graph-cuts from X-ray CT data1. Even though previously several researchers have used graph-cuts for image segmentation, to date, no systematic studies have been performed regarding the range of parameter that give accurate results. The energy function in the graph-cuts algorithm requires 3 suitable parameter settings: K, a large constant for assigning seed points, c, the similarity coefficient for n-links, and λ, the terminal coefficient for t-links. We analyzed the parameter sensitivity with four lung data sets from subjects with lung cancer using error metrics. Large values of K created artifacts on segmented images, and relatively much larger value of c than the value of λ influenced the balance between the boundary term and the data term in the energy function, leading to unacceptable segmentation results. For a range of parameter settings, we performed 3D image segmentation, and in each case compared the results with the expert-delineated lung boundaries. We used simple 6-neighborhood systems for n-link in 3D. The 3D image segmentation took 10 minutes for a 512x512x118 ~ 512x512x190 lung CT image volume. Our results indicate that the graph-cuts algorithm was more sensitive to the K and λ parameter settings than to the C parameter and furthermore that amongst the range of parameters tested, K=5 and λ=0.5 yielded good results.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.