6,479 Hits in 5.0 sec

Optimal MAP Parameters Estimation in STAPLE - Learning from Performance Parameters versus Image Similarity Information [chapter]

Subrahmanyam Gorthi, Alireza Akhondi-Asl, Jean-Philippe Thiran, Simon K. Warfield
2014 Lecture Notes in Computer Science  
; we also propose a methodology for transferring this prior knowledge about the performance parameters into the STAPLE algorithm through optimal setting of the MAP parameters.  ...  Simultaneous Truth And Performance Level Estimation (STAPLE) is a widely used fusion algorithm that simultaneously estimates both performance parameters for each template, and the output segmentation;  ...  In this paper, we introduce a general and powerful framework for learning prior knowledge about the performance parameters of each label in each template, and for using that information to optimally set  ... 
doi:10.1007/978-3-319-10581-9_22 fatcat:wdgknaqthverdc35coqukzxfni

Estimating A Reference Standard Segmentation With Spatially Varying Performance Parameters: Local MAP STAPLE

O. Commowick, A. Akhondi-Asl, S. K. Warfield
2012 IEEE Transactions on Medical Imaging  
Further, we propose an expression to compute confidence intervals in the estimated local performance parameters.  ...  We present a new algorithm, called local MAP STAPLE, to estimate from a set of multi-label segmentations both a reference standard segmentation and spatially varying performance parameters.  ...  ACKNOWLEDGMENTS This research was supported in part by NIH grants R01 RR021885, R01 EB008015, R03 EB008680, R01 LM010033 and R01 EB013248.  ... 
doi:10.1109/tmi.2012.2197406 pmid:22562727 pmcid:PMC3496174 fatcat:3qnuoovrzzdbbhpbp2jtxkcrae

Simultaneous Truth and Performance Level Estimation Through Fusion of Probabilistic Segmentations

Alireza Akhondi-Asl, Simon K. Warfield
2013 IEEE Transactions on Medical Imaging  
Intensity and label map images of each one of the aligned templates are used to train a local Gaussian mixture model based classifier.  ...  However, intensity weighted fusion approaches use local intensity similarity as a surrogate measure of local template quality for predicting target segmentation and do not seek to characterize template  ...  ACKNOWLEDGMENTS This research was supported in part by NIH grants R01 EB013248, R01 EB008015, R01 LM010033, R01 NS079788, U01 NS082320, R42 MH086984, P30 HD018655, and by a research grant from Boston Children's  ... 
doi:10.1109/tmi.2013.2266258 pmid:23744673 pmcid:PMC3788853 fatcat:igiglsxjmngwxkyip56xcafeti

Non-local statistical label fusion for multi-atlas segmentation

Andrew J. Asman, Bennett A. Landman
2013 Medical Image Analysis  
Despite success on human raters, current approaches inaccurately model multi-atlas behavior as they fail to seamlessly incorporate exogenous intensity information into the estimation process.  ...  NLS reformulates the STAPLE framework from a non-local means perspective in order to learn what label an atlas would have observed, given perfect correspondence.  ...  This work was conducted in part using the resources of the Advanced Computing Center for Research and Education (ACCRE) at Vanderbilt University, Nashville, TN. The authors are grateful to Dr.  ... 
doi:10.1016/ pmid:23265798 pmcid:PMC3648421 fatcat:x6sj5x3qmrdulanqgok2gqnumq

Whole-body bone segmentation from MRI for PET/MRI attenuation correction using shape-based averaging

Hossein Arabi, Habib Zaidi
2016 Medical Physics (Lancaster)  
In addition, the authors assessed the performance of simultaneous truth and performance level estimation (STAPLE) and the selective and iterative method for performance level estimation (SIMPLE) combined  ...  The SIMPLE method was applied globally (G-SIMPLE) while STAPLE method was employed at both global (G-STAPLE) and local (L-STAPLE) levels.  ...  Optimization of R window size for local atlas performance assessment in L-STAPLE method (top) and patch window size for local shape similarity assessment in the L-Shp method (bottom).  ... 
doi:10.1118/1.4963809 pmid:27806602 fatcat:of5knkavdrh57hd2dfcukuhfum

Formulating Spatially Varying Performance in the Statistical Fusion Framework

A. J. Asman, B. A. Landman
2012 IEEE Transactions on Medical Imaging  
STAPLE, globally weighted vote) or voxelwise (e.g. locally weighted vote) performance models.  ...  This approach, Spatial STAPLE, provides significant improvements over state-of-the-art label fusion algorithms in both simulated and empirical data sets.  ...  This work was supported in part by NIH/NINDS 1R01EB006136, NIH/NINDS 1R01EB006193, NIH/NINDS 1R03EB012461, and NIH/NINDS 1R21NS064534.  ... 
doi:10.1109/tmi.2012.2190992 pmid:22438513 pmcid:PMC3368083 fatcat:tgxlbhubbvhq5ajvsyz33tgv5q

Comparison of atlas-based techniques for whole-body bone segmentation

Hossein Arabi, Habib Zaidi
2017 Medical Image Analysis  
The local weighed atlas fusion approach using the MSD similarity measure outperformed the other strategies by achieving a DSC of 0.81 ± 0.03 while using the NCC and NMI measures resulted in a DSC of 0.78  ...  The performance evaluation of the different segmentation techniques was carried out in terms of estimating bone extraction accuracy from whole-body MRI using standard metrics, such as Dice similarity (  ...  The STAPLE algorithm estimates a ground truth bone map from given bone atlas binary maps ( T Sn ).  ... 
doi:10.1016/ pmid:27871000 fatcat:sye37orrpfaczaobfvrj4qqvqm

Estimation of the Prior Distribution of Ground Truth in the STAPLE Algorithm: An Empirical Bayesian Approach [chapter]

Alireza Akhondi-Asl, Simon K. Warfield
2012 Lecture Notes in Computer Science  
Our algorithm is a parametric empirical Bayesian extension of the STAPLE algorithm which uses the observations to accurately estimate the prior distribution of the hidden ground truth using an expectation  ...  We segment 128 principle gray and white matter structures of the brain using our novel method and eight other state-of-the-art algorithms in the literature.  ...  This investigation was supported in part by NIH grants R01 EB008015, R01 LM010033, R01 EB013248, and P30 HD018655 and by a research grant from the Boston Children's Hospital Translational Research Program  ... 
doi:10.1007/978-3-642-33415-3_73 fatcat:jmoviwolgzdwnjpvgdmf4hitjm

Multi-subject Registration for Unbiased Statistical Atlas Construction [chapter]

Mathieu De Craene, Aloys du Bois d'Aische, Benoît Macq, Simon K. Warfield
2004 Lecture Notes in Computer Science  
Our metric aligns each subject with a hidden probabilistic model of the common spatial distribution of anatomical tissues, estimated using STAPLE.  ...  The computational cost of joint simultaneous registration of the population of subjects is small due to the use of an efficient gradient estimate used to solve the optimization transform aligning each  ...  Terrie Inder for the neonate MRI data used in this study.  ... 
doi:10.1007/978-3-540-30135-6_80 fatcat:dbo2slm6frd5fk2547c3r4niie

Multi-Atlas Segmentation with Joint Label Fusion

Hongzhi Wang, J. W. Suh, S. R. Das, J. B. Pluta, C. Craige, P. A. Yushkevich
2013 IEEE Transactions on Pattern Analysis and Machine Intelligence  
This probability is approximated using intensity similarity between a pair of atlases and the target image in the neighborhood of each voxel.  ...  For instance, such errors can be reduced by optimally constructing a single atlas that is the most representative of the population using training data [12] , [11], [18].  ...  Acknowledgements We thank Sussane Mueller and Michael Weiner for providing the images used in our hippocampal subfield segmentation experiments.  ... 
doi:10.1109/tpami.2012.143 pmid:22732662 pmcid:PMC3864549 fatcat:kidcxtml7ngihd6hojfarlgfr4

Optimal weights for local multi-atlas fusion using supervised learning and dynamic information (SuperDyn): Validation on hippocampus segmentation

Ali R. Khan, Nicolas Cherbuin, Wei Wen, Kaarin J. Anstey, Perminder Sachdev, Mirza Faisal Beg
2011 NeuroImage  
We developed a novel method for spatially-local selection of atlas-weights in multi-atlas segmentation that combines supervised learning on a training set and dynamic information in the form of local registration  ...  accuracy estimates (SuperDyn).  ...  Registration accuracy estimates We used local estimates of registration accuracy to inform the dynamic component of weight selection, specifically the squared intensity difference of the images after registration  ... 
doi:10.1016/j.neuroimage.2011.01.078 pmid:21296166 fatcat:dqxbvl2lfbboth2x7j3fvdqhcq

A Generative Model for Image Segmentation Based on Label Fusion

Mert R Sabuncu, B T Thomas Yeo, K Van Leemput, Bruce Fischl, Polina Golland
2010 IEEE Transactions on Medical Imaging  
In a second experiment, we use brain MRI scans of 282 subjects to demonstrate that the proposed segmentation tool is sufficiently sensitive to robustly detect hippocampal volume changes in a study of aging  ...  In the first set of experiments, we use 39 brain MRI scans-with manually segmented white matter, cerebral cortex, ventricles and subcortical structures-to compare different label fusion algorithms and  ...  Furthermore, this difference can be due to the fact that the image intensities in the putamen and pallidum are similar to neighboring structures and thus local intensity information is less useful in segmenting  ... 
doi:10.1109/tmi.2010.2050897 pmid:20562040 pmcid:PMC3268159 fatcat:nt4bfbcc3ff5rjwhirdfviwqrq

Hierarchical performance estimation in the statistical label fusion framework

Andrew J. Asman, Bennett A. Landman
2014 Medical Image Analysis  
Label fusion is a critical step in many image segmentation frameworks (e.g., multi-atlas segmentation) as it provides a mechanism for generalizing a collection of labeled examples into a single estimate  ...  To address this problem, we propose a generalized statistical fusion framework using hierarchical models of rater performance.  ...  This research was conducted in part using the resources of the Advanced Computing Center for Research and Education at Vanderbilt University, Nashville, TN.  ... 
doi:10.1016/ pmid:25033470 pmcid:PMC4391396 fatcat:6n65duc6fbbwfkuuzdjcs3dn6e

A Supervised Patch-Based Approach for Human Brain Labeling

Françcois Rousseau, Piotr A. Habas, Colin Studholme
2011 IEEE Transactions on Medical Imaging  
Following recent developments in non-local image denoising, the similarity between images is represented by a weighted graph computed from an intensity-based distance between patches.  ...  Based on image intensity similarities between the input image and an anatomy textbook, an original strategy which does not require any non-rigid registration is presented.  ...  A pair-wise label propagation approach with the MV rule to fuse label maps has been used to determine the optimal parameter set.  ... 
doi:10.1109/tmi.2011.2156806 pmid:21606021 pmcid:PMC3318921 fatcat:7wlhkneop5da3d7tpjpv3bsdmu

Hippocampus Segmentation Based on Local Linear Mapping

Shumao Pang, Jun Jiang, Zhentai Lu, Xueli Li, Wei Yang, Meiyan Huang, Yu Zhang, Yanqiu Feng, Wenhua Huang, Qianjin Feng
2017 Scientific Reports  
In LLM, we assume that the MR and DF samples are located on two nonlinear manifolds and the mapping from the MR manifold to the DF manifold is differentiable and locally linear.  ...  We combine the MR dictionary using local linear representation to present the test sample, and combine the DF dictionary using the corresponding coefficients derived from local linear representation procedure  ...  The training dataset was used to construct the dictionary, and the optimization dataset was utilized to optimize parameters, which were used to perform the subsequent experiments using the test dataset  ... 
doi:10.1038/srep45501 pmid:28368016 pmcid:PMC5377362 fatcat:6lxzvgg27bd25b2urabr44mxhi
« Previous Showing results 1 — 15 out of 6,479 results