Challenges and Methodologies of Fully Automatic Whole Heart Segmentation: A Review

Zhuang, Xiahai

doi:10.1260/2040-2295.4.3.371

Cited by 141 publications

(84 citation statements)

References 118 publications

(323 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There is a wide range of multi-atlas based approaches addressing the problem of cardiac segmentation [7]. We have selected to use the segmentation pipeline proposed by Zuluaga et al [10] as it has shown to be robust in the segmentation of different structures within the heart [13,14].…”

Section: Methodsmentioning

confidence: 99%

“…While the segmentation of the main chambers of the heart has been widely addressed [7], little work has focused on the extraction of smaller structures such as the PM. Spreeuwers et al [8] were the first to tackle the problem through a 2D region-based approach.…”

Section: Introductionmentioning

confidence: 99%

“…More recently, Gao et al [9] addressed the problem through a topological method that tried to restore missing structures from an initial segmentation using high resolution CT. Other methods in the literature, have extracted the PM along with the left ventricle but, not as a separate structure. Despite the popularity of atlas-based methods in the segmentation of the heart [10,7], none of the existing works have tried to use this type of framework for the segmentation of the PM. This could be explained by the complexity, shape, size and position variability of these.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Papillary Muscle Segmentation from a Multi-atlas Database: A Feasibility Study

Biffi

Zuluaga

Ourselin

et al. 2016

Statistical Atlases and Computational Models of the Heart. Imaging and Modelling Challenges

View full text Add to dashboard Cite

Abstract. Mitral valve regurgitation replacement could be benefited by the use of pre-surgical 3D models that would allow the clinician to better understand the patient's structural and functional information before surgery. As no single image modality can provide all this information, it is necessary to fuse 3D transthoracic echo (TTE) and magnetic resonance imaging (MRI) to obtain a complete model. The registration of these two modalities, within the context of cardiac imaging, is a challenging task that requires manual intervention or the use of anatomical landmarks that can drive the registration. Within the context of mitral valve planning, the papillary muscles represent an ideal landmark set as they can be clearly identified in both modalities. However, to date little work has addressed the problem of papillary muscle segmentation. In this paper, we apply an atlas-based segmentation method for the automatic extraction of papillary muscles from MRI. Results show that a good quality segmentation (Dice score 0.67 ± 0.13) can be achieved within the straightforward pipeline provided by this approach. Hence, our atlas-based segmentation method could represent the first key step towards a novel TTE-MRI fusion algorithm.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Papillary Muscle Segmentation from a Multi-atlas Database: A Feasibility Study

Biffi

Zuluaga

Ourselin

et al. 2016

Statistical Atlases and Computational Models of the Heart. Imaging and Modelling Challenges

View full text Add to dashboard Cite

show abstract

“…For the LA+PVs segmentation, Dice score [30], Hausdorff distance (HD) [31] and Average Surface Distance (ASD) [30] were used as evaluation metrics. For the SAS, we employed leave-one-patient-out cross-validation (LOO CV) and reported the cross-validated accuracy, sensitivity, specificity, area under the receiver operating characteristic (ROC) curve (AUC), and the Dice score [32]- [34].…”

Section: Experimental Settings and Performance Measurementioning

confidence: 99%

A fully automatic deep learning method for atrial scarring segmentation from late gadolinium-enhanced MRI images

Yang

Zhuang

Khan

et al. 2017

2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017)

Self Cite

View full text Add to dashboard Cite

Precise and objective segmentation of atrial scarring (SAS) is a prerequisite for quantitative assessment of atrial fibrillation using non-invasive late gadolinium-enhanced (LGE) MRI. This also requires accurate delineation of the left atrium (LA) and pulmonary veins (PVs) geometry. Most previous studies have relied on manual segmentation of LA wall and PVs, which is a tedious and error-prone procedure with limited reproducibility. There are many attempts on automatic SAS using simple thresholding, histogram analysis, clustering and graph-cut based approaches; however, in general, these methods are considered as unsupervised learning thus subject to limited segmentation accuracy. In this study, we present a fully-automated multi-atlas based whole heart segmentation method to derive the LA and PVs geometry objectively that is followed by a fully automatic deep learning method for SAS. Our deep learning method consists of a feature extraction step via super-pixel over-segmentation and a supervised classification step via stacked sparse auto-encoders. We demonstrate the efficacy of our method on 20 clinical LGE MRI scans acquired from a longstanding persistent atrial fibrillation cohort. Both quantitative and qualitative results show that our fully automatic method obtained accurate segmentation results compared to the manual segmentation based ground truths. INTRODUCTIONAtrial fibrillation (AF) is the most common sustained heart rhythm disturbance encountered in adult cardiology. Several studies have shown that AF is correlated with electrical, contractile, and structural remodeling in the left atrium (LA) [1]. Moreover, LA fibrosis may be arrhythmogenic that causes more aggressive symptoms and makes difficulties in the management of AF [1]. Minimally invasive catheter ablation (CA) using radio-frequency energy has become one of the most common treatments for AF patients refractory to drug treatment [2]. CA aims to electrically isolate the pulmonary veins (PVs) from the left atrial (LA) body because previous studies show that ectopic beats from the PVs can frequently trigger the AF [3]. In this context, techniques have been developed to evaluate the LA wall composition and assess the circumferential PVs scarring that results from CA in order to understand the AF with proper management and prognosis. At present, electro-anatomical mapping (EAM) system, which is performed during the electrophysiological study, is considered to be a clinical reference standard technique for the assessment of the LA substrate and ablation-induced scarring. However, EAM is invasive and suffers from ionizing radiation and its suboptimal accuracy, which has reported errors of up to 10 mm in the localization of scar tissue [4]. Noninvasive late gadolinium-enhanced (LGE) MRI is an established method for visualizing and assessing myocardial infarction or fibrosis [5]. This is ascribed to the altered wash-in and wash-out contrast agent kinetics, and the hyper-enhancement reflects the increased interstitial space of the myocardium wit...

show abstract

“…Moreover, the vast literature in CMR analysis tools [14] proposes a large set of semi-and fully-automated methods for adult whole-heart segmentation that could be adapted and applied to paediatric CMR images to ease and improve the delineation and extraction of the heart.…”

Section: Introductionmentioning

confidence: 99%

Strengths and Pitfalls of Whole-Heart Atlas-Based Segmentation in Congenital Heart Disease Patients

Zuluaga

Biffi

Schievano

et al. 2017

Reconstruction, Segmentation, and Analysis of Medical Images

View full text Add to dashboard Cite

Abstract. Atlas-based whole-heart segmentation is a well-established technique for the extraction of key cardiac structures of the adult heart. Despite its relative success in this domain, its implementation in wholeheart segmentation of paediatric patients suffering from a form of congenital heart disease is not straightforward. The aim of this work is to evaluate the current strengths and limitations of whole-heart atlas based segmentation techniques within the context of the Whole-Heart and Great Vessel Segmentation from 3D Cardiovascular MRI in Congenital Heart Disease Challenge (HVSMR). Obtained results suggest that there are no significant differences in the accuracies of state-of-the-art methods, reporting maximum Dice scores of 0.73 for the myocardium and 0.90 for the blood pool.

show abstract

Challenges and Methodologies of Fully Automatic Whole Heart Segmentation: A Review

Cited by 141 publications

References 118 publications

Papillary Muscle Segmentation from a Multi-atlas Database: A Feasibility Study

Papillary Muscle Segmentation from a Multi-atlas Database: A Feasibility Study

A fully automatic deep learning method for atrial scarring segmentation from late gadolinium-enhanced MRI images

Strengths and Pitfalls of Whole-Heart Atlas-Based Segmentation in Congenital Heart Disease Patients

Contact Info

Product

Resources

About