Multilabel statistical shape prior for image segmentation

Grosgeorge, Damien; Petitjean, Caroline; Ruan, Su

doi:10.1049/iet-ipr.2015.0408

Cited by 7 publications

(8 citation statements)

References 41 publications

(55 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This occurs, for example, when there is manual selection of points [25,26,65]. Another option is deining additional vertices, called terminals, which are initially connected with all vertices of the graph and each one represents a class [6,36,46].…”

Section: Graph-based Approachesmentioning

confidence: 99%

“…For a pair of neighboring vertices representing pixels, the weight of the edge connecting them must relect the probability of the pair belonging to the same class. In this case, the similarity of pixel intensity is the most widely used factor [6,25,36,46,64,65,82]. On the edges connecting vertices representing pixels and terminal vertices, the weight relects the probability that the vertex belongs to the terminal class.…”

Section: Graph-based Approachesmentioning

confidence: 99%

See 1 more Smart Citation

Left Ventricle Segmentation in Cardiac MR: A Systematic Mapping of the Past Decade

Ribeiro

Nunes

2022

ACM Comput. Surv.

View full text Add to dashboard Cite

Left ventricle segmentation in short-axis cardiac magnetic resonance images is important to diagnose heart disease. However, the repetitive manual segmentation of these images requires considerable human effort and can decrease diagnostic accuracy. In recent years, several fully and semi-automatic approaches have been proposed, mainly using image-based, atlas, graphs, deformable models, and artificial intelligence methods. This paper presents a systematic mapping on the left ventricle segmentation, considering 74 studies published in the last decade. The main contributions of this review are: definition of the main segmentation challenges in these images; proposal of a new schematization, dividing the segmentation process into stages; categorization and analysis of the segmentation methods, including hybrid combinations; and analysis of the evaluation process, metrics, and databases. The performance of the methods in the most used public database is assessed, and the main limitations, weaknesses, and strengths of each method category are presented. Finally, trends, challenges, and research opportunities are discussed. The analysis indicates that methods from all categories can achieve good performance, and hybrid methods combining deep learning and deformable models obtain the best results. Methods still fail in specific slices, segment wrong regions, and produce anatomically impossible segmentations.

show abstract

Section: Graph-based Approachesmentioning

confidence: 99%

Section: Graph-based Approachesmentioning

confidence: 99%

Left Ventricle Segmentation in Cardiac MR: A Systematic Mapping of the Past Decade

Ribeiro

Nunes

2022

ACM Comput. Surv.

View full text Add to dashboard Cite

show abstract

“…In this context, plenty of work has been developed prior to the advent of deep learning for segmentation given variational approaches. Multi-atlas based approaches and machine learning based techniques were also ways to integrate prior information through relying on labeled data [2]. However, there is no straightforward way to transfer these previous works to deep learning networks, since the latter have some specific constraints including differentiability and optimization of the loss function.…”

Section: Introductionmentioning

confidence: 99%

Investigating CoordConv for Fully and Weakly Supervised Medical Image Segmentation

Jurdi

Dargent

Petitjean

et al. 2020

2020 Tenth International Conference on Image Processing Theory, Tools and Applications (IPTA)

Self Cite

View full text Add to dashboard Cite

Convolutional neural networks (CNN) have established state-of-the-art performance in computer vision tasks such as object detection and segmentation. One of the major remaining challenges concerns their ability to capture consistent spatial attributes, especially in medical image segmentation. A way to address this issue is through integrating localization prior into system architecture. The CoordConv layers are extensions of convolutional neural network wherein convolution is conditioned on spatial coordinates. This paper investigates CoordConv as a proficient substitute to convolutional layers for organ segmentation in both fully and weakly supervised settings. Experiments are conducted on two public datasets, SegTHOR, which focuses on the segmentation of thoracic organs at risk in computed tomography (CT) images, and ACDC, which addresses ventricular endocardium segmentation of the heart in MR images. We show that if CoordConv does not significantly increase the accuracy with respect to standard convolution, it may interestingly increase model convergence at almost no additional computational cost.

show abstract

“…Fully automatic methods in [5], [6] consist in first finding the approximate shape using a detect and connect approach, and then a classifier is trained to find short segments of the esophagus which are approximated by an elliptical model. In [7] the approach makes use of a training set of manually contoured images to build a shape prior, which is incorporated into a graph cut framework. However the shape model was unidimensional and not fully leveraging the training set.…”

Section: Introductionmentioning

confidence: 99%

Fully automated esophagus segmentation with a hierarchical deep learning approach

Trullo

Petitjean

Nie

et al. 2017

2017 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)

Self Cite

View full text Add to dashboard Cite

Segmentation of organs at risk in CT volumes is a prerequisite for radiotherapy treatment planning. In this paper we focus on esophagus segmentation, a challenging problem since the walls of the esophagus have a very low contrast in CT images. Making use of Fully Convolutional Networks (FCN), we present several extensions that improve the performance, including a new architecture that allows to use low level features with high level information, effectively combining local and global information for improving the localization accuracy. Experiments demonstrate competitive performance on a dataset of 30 CT scans.

show abstract

Multilabel statistical shape prior for image segmentation

Cited by 7 publications

References 41 publications

Left Ventricle Segmentation in Cardiac MR: A Systematic Mapping of the Past Decade

Left Ventricle Segmentation in Cardiac MR: A Systematic Mapping of the Past Decade

Investigating CoordConv for Fully and Weakly Supervised Medical Image Segmentation

Fully automated esophagus segmentation with a hierarchical deep learning approach

Contact Info

Product

Resources

About