A topo-graph model for indistinct target boundary definition from anatomical images

Cui, H; Wang, X; Zhou, J; Gong, G; Eberl, S; Yin, Y; Wang, L; Feng, D; Fulham, M

A topo-graph model for indistinct target boundary definition from anatomical images

Cui, H Wang, X Zhou, J

Gong, G Eberl, S Yin, Y Wang, L Feng, D Fulham, M

Permalink

Publication Type:: Journal Article
Citation:: Computer Methods and Programs in Biomedicine, 2018, 159 pp. 211 - 222
Issue Date:: 2018-06-01

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download Accepted Manuscript VersionAdobe PDF (1.39 MB)

View on publisher's site

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Cui, H	en_US
dc.contributor.author	Wang, X	en_US
dc.contributor.author	Zhou, J https://orcid.org/0000-0001-6034-644X	en_US
dc.contributor.author	Gong, G	en_US
dc.contributor.author	Eberl, S	en_US
dc.contributor.author	Yin, Y	en_US
dc.contributor.author	Wang, L	en_US
dc.contributor.author	Feng, D	en_US
dc.contributor.author	Fulham, M	en_US
dc.date.available	2020-05-25T19:04:46Z
dc.date.issued	2018-06-01	en_US
dc.identifier.citation	Computer Methods and Programs in Biomedicine, 2018, 159 pp. 211 - 222	en_US
dc.identifier.issn	0169-2607	en_US
dc.identifier.uri	http://hdl.handle.net/10453/131803
dc.description.abstract	© 2018 Elsevier B.V. Background and Objective: It can be challenging to delineate the target object in anatomical imaging when the object boundaries are difficult to discern due to the low contrast or overlapping intensity distributions from adjacent tissues. Methods: We propose a topo-graph model to address this issue. The first step is to extract a topographic representation that reflects multiple levels of topographic information in an input image. We then define two types of node connections - nesting branches (NBs) and geodesic edges (GEs). NBs connect nodes corresponding to initial topographic regions and GEs link the nodes at a detailed level. The weights for NBs are defined to measure the similarity of regional appearance, and weights for GEs are defined with geodesic and local constraints. NBs contribute to the separation of topographic regions and the GEs assist the delineation of uncertain boundaries. Final segmentation is achieved by calculating the relevance of the unlabeled nodes to the labels by the optimization of a graph-based energy function. We test our model on 47 low contrast CT studies of patients with non-small cell lung cancer (NSCLC), 10 contrast-enhanced CT liver cases and 50 breast and abdominal ultrasound images. The validation criteria are the Dice's similarity coefficient and the Hausdorff distance. Results: Student's t-test show that our model outperformed the graph models with pixel-only, pixel and regional, neighboring and radial connections (p-values <0.05). Conclusions: Our findings show that the topographic representation and topo-graph model provides improved delineation and separation of objects from adjacent tissues compared to the tested models.	en_US
dc.relation.ispartof	Computer Methods and Programs in Biomedicine	en_US
dc.relation.isbasedon	10.1016/j.cmpb.2018.03.018	en_US
dc.subject.classification	Medical Informatics	en_US
dc.subject.mesh	Abdomen	en_US
dc.subject.mesh	Breast	en_US
dc.subject.mesh	Liver	en_US
dc.subject.mesh	Humans	en_US
dc.subject.mesh	Carcinoma, Non-Small-Cell Lung	en_US
dc.subject.mesh	Lung Neoplasms	en_US
dc.subject.mesh	Contrast Media	en_US
dc.subject.mesh	Tomography, X-Ray Computed	en_US
dc.subject.mesh	Ultrasonography	en_US
dc.subject.mesh	Cluster Analysis	en_US
dc.subject.mesh	Models, Statistical	en_US
dc.subject.mesh	Sensitivity and Specificity	en_US
dc.subject.mesh	Reproducibility of Results	en_US
dc.subject.mesh	Algorithms	en_US
dc.subject.mesh	Image Processing, Computer-Assisted	en_US
dc.subject.mesh	Female	en_US
dc.title	A topo-graph model for indistinct target boundary definition from anatomical images	en_US
dc.type	Journal Article
utslib.citation.volume	159	en_US
utslib.for	0903 Biomedical Engineering	en_US
utslib.for	0801 Artificial Intelligence and Image Processing	en_US
utslib.for	0906 Electrical and Electronic Engineering	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
utslib.copyright.status	open_access
pubs.publication-status	Published	en_US
pubs.volume	159	en_US

Abstract:

© 2018 Elsevier B.V. Background and Objective: It can be challenging to delineate the target object in anatomical imaging when the object boundaries are difficult to discern due to the low contrast or overlapping intensity distributions from adjacent tissues. Methods: We propose a topo-graph model to address this issue. The first step is to extract a topographic representation that reflects multiple levels of topographic information in an input image. We then define two types of node connections - nesting branches (NBs) and geodesic edges (GEs). NBs connect nodes corresponding to initial topographic regions and GEs link the nodes at a detailed level. The weights for NBs are defined to measure the similarity of regional appearance, and weights for GEs are defined with geodesic and local constraints. NBs contribute to the separation of topographic regions and the GEs assist the delineation of uncertain boundaries. Final segmentation is achieved by calculating the relevance of the unlabeled nodes to the labels by the optimization of a graph-based energy function. We test our model on 47 low contrast CT studies of patients with non-small cell lung cancer (NSCLC), 10 contrast-enhanced CT liver cases and 50 breast and abdominal ultrasound images. The validation criteria are the Dice's similarity coefficient and the Hausdorff distance. Results: Student's t-test show that our model outperformed the graph models with pixel-only, pixel and regional, neighboring and radial connections (p-values <0.05). Conclusions: Our findings show that the topographic representation and topo-graph model provides improved delineation and separation of objects from adjacent tissues compared to the tested models.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/131803