Fully automated segmentation of the left atrium, pulmonary veins, and left atrial appendage from magnetic resonance angiography by joint-atlas-optimization.

PURPOSE: Atrial fibrillation (AF) originating from the left atrium (LA) and pulmonary veins (PVs) is the most prevalent cardiac electrophysiological disorder. Accurate segmentation and quantification of the LA chamber, PVs, and left atrial appendage (LAA) provides clinically important references for treatment of AF patients. The purpose of this work is to realize objective segmentation of the LA chamber, PVs, and LAA in an accurate and fully automated manner. METHODS: In this work, we proposed a new approach, named joint-atlas-optimization, to segment the LA chamber, PVs, and LAA from magnetic resonance angiography (MRA) images. We formulated the segmentation as a single registration problem between the given image and all N atlas images, instead of N separate registration between the given image and an individual atlas image. Level sets was applied to refine the atlas-based segmentation. Using the publically available LA benchmark database, we compared the proposed joint-atlas-optimization approach to the conventional pairwise atlas approach and evaluated the segmentation performance in terms of Dice index and surface-to-surface (S2S) distance to the manual ground truth. RESULTS: The proposed joint-atlas-optimization method showed systemically improved accuracy and robustness over the pairwise atlas approach. The Dice of LA segmentation using joint-atlas-optimization was 0.93 ± 0.04, compared to 0.91 ± 0.04 by the pairwise approach (P < 0.05). The mean S2S distance was 1.52 ± 0.58 mm, compared to 1.83 ± 0.75 mm (P < 0.05). In particular, it produced significantly improved segmentation accuracy of the LAA and PVs, the small distant part in LA geometry that is intrinsically difficult to segment using the conventional pairwise approach. The Dice of PVs segmentation was 0.69 ± 0.16, compared to 0.49 ± 0.15 (P < 0.001). The Dice of LAA segmentation was 0.91 ± 0.03, compared to 0.88 ± 0.05 (P < 0.01). CONCLUSION: The proposed joint-atlas optimization method can segment the complex LA geometry in a fully automated manner. Compared to the conventional atlas approach in a pairwise manner, our method improves the performance on small distal parts of LA, for example, PVs and LAA, the geometrical and quantitative assessment of which is clinically interesting.

A deep learning framework for unsupervised affine and deformable image registration.

Image registration, the process of aligning two or more images, is the core technique of many (semi-)automatic medical image analysis tasks. Recent studies have shown that deep learning methods, notably convolutional neural networks (ConvNets), can be used for image registration. Thus far training of ConvNets for registration was supervised using predefined example registrations. However, obtaining example registrations is not trivial. To circumvent the need for predefined examples, and thereby to increase convenience of training ConvNets for image registration, we propose the Deep Learning Image Registration (DLIR) framework for unsupervised affine and deformable image registration. In the DLIR framework ConvNets are trained for image registration by exploiting image similarity analogous to conventional intensity-based image registration. After a ConvNet has been trained with the DLIR framework, it can be used to register pairs of unseen images in one shot. We propose flexible ConvNets designs for affine image registration and for deformable image registration. By stacking multiple of these ConvNets into a larger architecture, we are able to perform coarse-to-fine image registration. We show for registration of cardiac cine MRI and registration of chest CT that performance of the DLIR framework is comparable to conventional image registration while being several orders of magnitude faster.

Robust motion correction for myocardial T1 and extracellular volume mapping by principle component analysis-based groupwise image registration.

BACKGROUND: Myocardial tissue characterization by MR T1 and extracellular volume (ECV) mapping has demonstrated clinical value. The modified Look-Locker inversion recovery (MOLLI) sequence is a standard mapping technique, but its quality can be negatively affected by motion. PURPOSE: To develop a robust motion correction method for T1 and ECV mapping. STUDY TYPE: Retrospective analysis of clinical data. POPULATION: Fifty patients who were referred to cardiac MR exam for T1 mapping. FIELD STRENGTH/SEQUENCE: 3.0T cardiac MRI with precontrast and postcontrast MOLLI acquisition of the left ventricle (LV). ASSESSMENT: A groupwise registration method based on principle component analysis (PCA) was developed to register all MOLLI frames simultaneously. The resulting T1 and ECV maps were compared to those from the original and motion-corrected MOLLI with pairwise registration, in terms of standard deviation (SD) error. STATISTICAL TEST: Paired variables were compared using the Wilcoxon signed-rank test. RESULTS: The groupwise registration method demonstrated improved registration performance compared to pairwise registration, with the T1 SD error reduced from 31 ± 20 msec to 26 ± 15 msec (P < 0.05), and ECV SD error reduced from 4.1 ± 3.6% to 2.8 ± 2.0% (P < 0.05). In LV segmental analysis, the performance was particularly improved in lateral segments, which are most affected by motion. The running time of groupwise registration was significantly shorter than that of the pairwise registration, 17.5 ± 3.0 seconds compared to 43.5 ± 2.2 seconds (P < 0.05). DATA CONCLUSION: We developed an automatic, robust motion correction method for myocardial T1 and ECV mapping based on a new groupwise registration scheme. The method led to lower mapping error compared to the conventional pairwise registration method in reduced execution time. LEVEL OF EVIDENCE: 3 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2018;47:1397-1405.

Fully automatic segmentation of left atrium and pulmonary veins in late gadolinium-enhanced MRI: Towards objective atrial scar assessment.

PURPOSE: To realize objective atrial scar assessment, this study aimed to develop a fully automatic method to segment the left atrium (LA) and pulmonary veins (PV) from late gadolinium-enhanced (LGE) magnetic resonance imaging (MRI). The extent and distribution of atrial scar, visualized by LGE-MRI, provides important information for clinical treatment of atrial fibrillation (AF) patients. MATERIALS AND METHODS: Forty-six AF patients (age 62 ± 8, 14 female) who underwent cardiac MRI prior to RF ablation were included. A contrast-enhanced MR angiography (MRA) sequence was acquired for anatomy assessment followed by an LGE sequence for LA scar assessment. A fully automatic segmentation method was proposed consisting of two stages: 1) global segmentation by multiatlas registration; and 2) local refinement by 3D level-set. These automatic segmentation results were compared with manual segmentation. RESULTS: The LA and PVs were automatically segmented in all subjects. Compared with manual segmentation, the method yielded a surface-to-surface distance of 1.49 ± 0.65 mm in the LA region when using both MRA and LGE, and 1.80 ± 0.93 mm when using LGE alone (P < 0.05). In the PV regions, the distance was 2.13 ± 0.67 mm and 2.46 ± 1.81 mm (P < 0.05), respectively. The difference between automatic and manual segmentation was comparable to the interobserver difference (P = 0.8 in LA region and P = 0.7 in PV region). CONCLUSION: We developed a fully automatic method for LA and PV segmentation from LGE-MRI, with comparable performance to a human observer. Inclusion of an MRA sequence further improves the segmentation accuracy. The method leads to automatic generation of a patient-specific model, and potentially enables objective atrial scar assessment for AF patients. J. Magn. Reson. Imaging 2016;44:346-354.

Multiatlas-based segmentation with preregistration atlas selection.

PURPOSE: Automatic, atlas-based segmentation of medical images benefits from using multiple atlases, mainly in terms of robustness. However, a large disadvantage of using multiple atlases is the large computation time that is involved in registering atlas images to the target image. This paper aims to reduce the computation load of multiatlas-based segmentation by heuristically selecting atlases before registration. METHODS: To be able to select atlases, pairwise registrations are performed for all atlas combinations. Based on the results of these registrations, atlases are clustered, such that each cluster contains atlas that registers well to each other. This can all be done in a preprocessing step. Then, the representatives of each cluster are registered to the target image. The quality of the result of this registration is estimated for each of the representatives and used to decide which clusters to fully register to the target image. Finally, the segmentations of the registered images are combined into a single segmentation in a label fusion procedure. RESULTS: The authors perform multiatlas segmentation once with postregistration atlas selection and once with the proposed preregistration method, using a set of 182 segmented atlases of prostate cancer patients. The authors performed the full set of 182 leave-one-out experiments and in each experiment compared the result of the atlas-based segmentation procedure to the known segmentation of the atlas that was chosen as a target image. The results show that preregistration atlas selection is slightly less accurate than postregistration atlas selection, but this is not statistically significant. CONCLUSIONS: Based on the results the authors conclude that the proposed method is able to reduce the number of atlases that have to be registered to the target image with 80% on average, without compromising segmentation accuracy.