Limits...
Segmentation of Pulmonary Vascular Trees from Thoracic 3D CT Images.

Shikata H, McLennan G, Hoffman EA, Sonka M - Int J Biomed Imaging (2009)

Bottom Line: A quantitative validation was performed with more than 1000 manually identified points selected from inside the vessel segments to assess true positives (TPs) and 1000 points randomly placed outside of the vessels to evaluate false positives (FPs) in each case.On average, for both the high and low volume lung images, 99% of the points was properly marked as vessel and 1% of the points were assessed as FPs.Our hybrid segmentation algorithm provides a highly reliable method of segmenting the combined pulmonary venous and arterial trees which in turn will serve as a critical starting point for further quantitative analysis tasks and aid in our overall goal of establishing a normative atlas of the human lung.

View Article: PubMed Central - PubMed

Affiliation: Iowa Institute for Biomedical Imaging, The University of Iowa, Iowa City, IA 52242, USA.

ABSTRACT
This paper describes an algorithm for extracting pulmonary vascular trees (arteries plus veins) from three-dimensional (3D) thoracic computed tomographic (CT) images. The algorithm integrates tube enhancement filter and traversal approaches which are based on eigenvalues and eigenvectors of a Hessian matrix to extract thin peripheral segments as well as thick vessels close to the lung hilum. The resultant algorithm was applied to a simulation data set and 44 scans from 22 human subjects imaged via multidetector-row CT (MDCT) during breath holds at 85% and 20% of their vital capacity. A quantitative validation was performed with more than 1000 manually identified points selected from inside the vessel segments to assess true positives (TPs) and 1000 points randomly placed outside of the vessels to evaluate false positives (FPs) in each case. On average, for both the high and low volume lung images, 99% of the points was properly marked as vessel and 1% of the points were assessed as FPs. Our hybrid segmentation algorithm provides a highly reliable method of segmenting the combined pulmonary venous and arterial trees which in turn will serve as a critical starting point for further quantitative analysis tasks and aid in our overall goal of establishing a normative atlas of the human lung.

No MeSH data available.


Related in: MedlinePlus

Eight distinct regions were delineated based upon their radial distance from the hilum. Two points were manually placed in each lung as landmarks denoting the hilum and the lung was three-dimensionally divided into a total of eight regions depending on the distance from the hilar points. Note that this interaction step is only used for the validation purposes.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2801012&req=5

fig6: Eight distinct regions were delineated based upon their radial distance from the hilum. Two points were manually placed in each lung as landmarks denoting the hilum and the lung was three-dimensionally divided into a total of eight regions depending on the distance from the hilar points. Note that this interaction step is only used for the validation purposes.

Mentions: In order to evaluate the segmentation results quantitatively, more than 1000 points in the vessels were manually identified in each CT data set to form a validation set to assess the true positive (TP) rate. The points were defined by an experienced observer trained and supervised by a pulmonologist. TP rate is defined as the ratio between the total number of points detected by the algorithm and the total number of points in the data set. We consistently chose the center point of the vessels as a member of the TP point set because the center line is more important than vessel borders in defining the vessel geometry. Each lung was divided into four distinct regions in terms of radial distance from the hilum so as to avoid biased distribution of the test points. At least 100 points were identified in each region. Figure 6 shows the four regions in each lung. Region A is the closest to the hilum and region D is close to the pleural surface. Region A contains a greater number of thicker segments than the other regions.


Segmentation of Pulmonary Vascular Trees from Thoracic 3D CT Images.

Shikata H, McLennan G, Hoffman EA, Sonka M - Int J Biomed Imaging (2009)

Eight distinct regions were delineated based upon their radial distance from the hilum. Two points were manually placed in each lung as landmarks denoting the hilum and the lung was three-dimensionally divided into a total of eight regions depending on the distance from the hilar points. Note that this interaction step is only used for the validation purposes.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2801012&req=5

fig6: Eight distinct regions were delineated based upon their radial distance from the hilum. Two points were manually placed in each lung as landmarks denoting the hilum and the lung was three-dimensionally divided into a total of eight regions depending on the distance from the hilar points. Note that this interaction step is only used for the validation purposes.
Mentions: In order to evaluate the segmentation results quantitatively, more than 1000 points in the vessels were manually identified in each CT data set to form a validation set to assess the true positive (TP) rate. The points were defined by an experienced observer trained and supervised by a pulmonologist. TP rate is defined as the ratio between the total number of points detected by the algorithm and the total number of points in the data set. We consistently chose the center point of the vessels as a member of the TP point set because the center line is more important than vessel borders in defining the vessel geometry. Each lung was divided into four distinct regions in terms of radial distance from the hilum so as to avoid biased distribution of the test points. At least 100 points were identified in each region. Figure 6 shows the four regions in each lung. Region A is the closest to the hilum and region D is close to the pleural surface. Region A contains a greater number of thicker segments than the other regions.

Bottom Line: A quantitative validation was performed with more than 1000 manually identified points selected from inside the vessel segments to assess true positives (TPs) and 1000 points randomly placed outside of the vessels to evaluate false positives (FPs) in each case.On average, for both the high and low volume lung images, 99% of the points was properly marked as vessel and 1% of the points were assessed as FPs.Our hybrid segmentation algorithm provides a highly reliable method of segmenting the combined pulmonary venous and arterial trees which in turn will serve as a critical starting point for further quantitative analysis tasks and aid in our overall goal of establishing a normative atlas of the human lung.

View Article: PubMed Central - PubMed

Affiliation: Iowa Institute for Biomedical Imaging, The University of Iowa, Iowa City, IA 52242, USA.

ABSTRACT
This paper describes an algorithm for extracting pulmonary vascular trees (arteries plus veins) from three-dimensional (3D) thoracic computed tomographic (CT) images. The algorithm integrates tube enhancement filter and traversal approaches which are based on eigenvalues and eigenvectors of a Hessian matrix to extract thin peripheral segments as well as thick vessels close to the lung hilum. The resultant algorithm was applied to a simulation data set and 44 scans from 22 human subjects imaged via multidetector-row CT (MDCT) during breath holds at 85% and 20% of their vital capacity. A quantitative validation was performed with more than 1000 manually identified points selected from inside the vessel segments to assess true positives (TPs) and 1000 points randomly placed outside of the vessels to evaluate false positives (FPs) in each case. On average, for both the high and low volume lung images, 99% of the points was properly marked as vessel and 1% of the points were assessed as FPs. Our hybrid segmentation algorithm provides a highly reliable method of segmenting the combined pulmonary venous and arterial trees which in turn will serve as a critical starting point for further quantitative analysis tasks and aid in our overall goal of establishing a normative atlas of the human lung.

No MeSH data available.


Related in: MedlinePlus