Multi-Atlas Segmentation with Joint Label Fusion

By Hongzhi Wang, J. Suh, Sandhitsu R. Das, John Pluta, C. Craige, Paul Yushkevich. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012

854 citations recorded by the source corpus.

Multi-atlas segmentation is an effective approach for automatically labeling objects of interest in biomedical images. In this approach, multiple expert-segmented example images, called atlases, are registered to a target image, and deformed atlas segmentations are combined using label fusion. Among the proposed label fusion strategies, weighted voting with spatially varying weight distributions derived from atlas-target intensity similarity have been particularly successful. However, one limitation of these strategies is that the weights are computed independently for each atlas, without taking into account the fact that different atlases may produce similar label errors. To address this limitation, we propose a new solution for the label fusion problem in which weighted voting is formulated in terms of minimizing the total expectation of labeling error and in which pairwise dependency between atlases is explicitly modeled as the joint probability of two atlases making a segmentation error at a voxel. This probability is approximated using intensity similarity between a pair of atlases and the target image in the neighborhood of each voxel. We validate our method in two medical image segmentation problems: hippocampus segmentation and hippocampus subfield segmentation in magnetic resonance (MR) images. For both problems, we show consistent and significant improvement over label fusion strategies that assign atlas weights independently.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/2dfebc4e812ff2b8dd97a412bee0e97d7be5210f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Deep convolutional networks for automated detection of posterior-element fractures on spine CT (2016, methodology): “First, the vertebra bodies of the spine with its posterior elements are segmented in spine CT using multi-atlas label fusion.(10,11) A set of atlases of the vertebra bodies are registered to the target image with free-form deformation.”
VoxResNet: Deep voxelwise residual networks for brain segmentation from 3D MR images (2017, neutral): “For example, multi-atlas label fusion (MALF) made use of multiple reference atlases and achieved good performance in brain segmentation tasks (Aljabar et al., 2009; Lötjönen et al., 2010; Wang et al., 2013; Heckemann et al., 2006).”
Data Augmentation Using Learned Transformations for One-Shot Medical Image Segmentation (2019, background): “When multiple atlases are available, they are each aligned to a target volume, and the warped atlas labels are fused [36, 41, 68, 78].”
Deep Learning and Convolutional Neural Networks for Medical Image Computing (2017, methodology): “Such systemsmay reduce errors, but could also cause disruption in the healthcare industry [46].”
Large-scale evaluation of ANTs and FreeSurfer cortical thickness measurements (2014, methodology): “The propagation of the DKT labels to each subject using label fusion as described earlier was performed in parallel and took anywhere between 40 and 80 h per subject for 16 serial image registrations and the application of the joint label fusion algorithm (Wang et al., 2013).”
Automatic Multi-organ Segmentation on Abdominal CT with Dense V-networks (2018, methodology): “DEEDS and JLF computations were performed using the publicly available deedsRegSSC (http://www.mpheinrich.de/software.html) and PICSL Multi-Atlas Segmentation Tool (https:// www.nitrc.org/projects/picsl_malf) implementations, respectively, using published abdominal CT registration parameters for DEE”
A review on automatic fetal and neonatal brain MRI segmentation (2017, neutral): “…et al. (2007) Shi et al. (2010) Cardoso et al. (2011) Shi et al. (2011a) Ledig et al. (2012) Makropoulos et al. (2012b) Melbourne et al. (2012) Wang et al. (2012c) Wu and Avants (2012) Cardoso et al. (2013) Beare et al. (2016) Liu et al. (2016) Classification Anbeek et al.…”
Automated Volumetry and Regional Thickness Analysis of Hippocampal Subfields and Medial Temporal Cortical Structures in Mild Cognitive Impairment (2014, methodology): “The CL classifiers contained in the atlas package are applied to the consensus segmentation produced by JLF.”
A bi-ventricular cardiac atlas built from 1000+ high resolution MR images of healthy subjects and an analysis of shape and motion (2015, neutral): “Multi-atlas segmentation has been demonstrated as a robust and accurate segmentation method for brain and cardiac images in the recent years (Asman and Landman, 2013; Bai et al., 2013; Wang et al., 2013).”
Spatial aggregation of holistically‐nested convolutional neural networks for automated pancreas localization and segmentation☆ (2017, methodology): “This type of organ segmentation strategy is widely used for many organ segmentation problems, such as the brain [11], [13], heart [12], lung [14], and pancreas [3], [2], [4], [5].”
Local Temperature Scaling for Probability Calibration (2020, methodology): “The Best Fusion bound, which is obtained by assigning the correct label to the segmentation result if at least one atlas provides the right label; and the Best Calibration bound, which is obtained by assigning a probability of 1 if the prediction by VoteNet+ is correct and 1/|L| otherwise, followed ”
MUSE: MUlti-atlas region Segmentation utilizing Ensembles of registration algorithms and parameters, and locally optimal atlas selection (2015, positive): “, 2014), or local similarity based weighting in label fusion (Artaechevarria et al., 2009; Isgum et al., 2009; Khan et al., 2011; Wang et al., 2012) were proposed to address these challenges, by selecting, either locally or globally, warps most similar to the target image.”
Harmonization of cortical thickness measurements across scanners and sites (2017, methodology): “The atlases and labels were warped to subject space and input to joint label fusion (Wang et al., 2013).”