ShapeCut: Bayesian surface estimation using shape-driven graph

Abstract

A variety of medical image segmentation problems present significant technical challenges, including heterogeneous pixel intensities, noisy/ill-defined boundaries and irregular shapes with high variability. The strategy of estimating optimal segmentations within a statistical framework that combines image data with priors on anatomical structures promises to address some of these technical challenges. However, methods that rely on local optimization techniques and/or local shape penalties (e.g., smoothness) have been proven to be inadequate for many difficult segmentation problems. These challenging segmentation problems can benefit from the inclusion of global shape priors within a maximum-a-posteriori estimation framework, which biases solutions toward an object class of interest. In this paper, we propose a maximum-a-posteriori formulation that relies on a generative image model by incorporating both local and global shape priors. The proposed method relies on graph cuts as well as a new shape parameters estimation that provides a global updates-based optimization strategy. We demonstrate our approach on synthetic datasets as well as on the left atrial wall segmentation from late-gadolinium enhancement MRI, which has been shown to be effective for identifying myocardial fibrosis in the diagnosis of atrial fibrillation. Experimental results prove the effectiveness of the proposed approach in terms of the average surface distance between extracted surfaces and the corresponding ground-truth, as well as the clinical efficacy of the method in the identification of fibrosis and scars in the atrial wall.

Fully Automatic Left Atrium Segmentation From Late Gadolinium Enhanced Magnetic Resonance Imaging Using a Dual Fully Convolutional Neural Network

“…Currently, our study is among the few that have attempted at direct automatic segmentation of the LA from LGE-MRIs [14,46,47]. Out of all existing attempts, most of the LGE-MRI studies, either in-vivo or ex-vivo, have relied heavily on manual segmentation [9,[48][49][50].…”

“…A benchmark study published by Tobon-Gomez et al compared the performance of nine different algorithms for LA segmentation from non-gadolinium enhanced MRIs/CT and showed that methodologies combining statistical models with regional growing approaches were the most effective [13]. Similar techniques have also been proposed and further improved upon for segmenting the LA from LGE-MRIs in studies by Veni et al [14], Zhu et al [15] and Tao et al [16]. Despite these recent efforts, most of the existing structural analysis studies, especially those that utilize clinical LGE-MRIs, are still based on laborintensive and error-prone manual segmentation approaches [9,10,12].…”

A global benchmark of algorithms for segmenting the left atrium from late gadolinium-enhanced cardiac magnetic resonance imaging

“…However, due to the limited number of submitted algorithms, the study was not able to draw any definitive conclusions in terms of approach development. Other studies on LGE-MRI segmentation also have limited efficacy as the methods proposed require additional information such as manually initialized shape priors (Veni, Elhabian andWhitaker 2017, Zhu et al 2013) or paired magnetic resonance angiography (Tao et al 2016). While LGE-MRI segmentation still heavily relies on traditional methods, many recent advancements in CNNs have been made for image segmentation in general.…”

Atrial scar quantification via multi-scale CNN in the graph-cuts framework

“…The boundary term in these works was normally defined according to the dissimilarity of intensity and distance between two connected nodes. Veni et al (2017) designed a regional term based on a generative image model incorporating both local and global shape priors. The boundary term was defined for regularizing the smoothness of the estimated surface, i.e., minimizing the squared difference of the offsets between neighboring vertices.…”