A unified framework for clustering and quantitative analysis of white matter fiber tracts. Maddah, M., Grimson, W., Warfield, S., & Wells, W. Med Image Anal, 04, 2008. abstract bibtex We present a novel approach for joint clustering and point-by-point mapping of white matter fiber pathways. Knowledge of the point correspondence along the fiber pathways is not only necessary for accurate clustering of the trajectories into fiber bundles, but also crucial for any tract-oriented quantitative analysis. We employ an expectation-maximization (EM) algorithm to cluster the trajectories in a gamma mixture model context. The result of clustering is the probabilistic assignment of the fiber trajectories to each cluster, an estimate of the cluster parameters, i.e. spatial mean and variance, and point correspondences. The fiber bundles are modeled by the mean trajectory and its spatial variation. Point-by-point correspondence of the trajectories within a bundle is obtained by constructing a distance map and a label map from each cluster center at every iteration of the EM algorithm. This offers a time-efficient alternative to pairwise curve matching of all trajectories with respect to each cluster center. The proposed method has the potential to benefit from an anatomical atlas of fiber tracts by incorporating it as prior information in the EM algorithm. The algorithm is also capable of handling outliers in a principled way. The presented results confirm the efficiency and effectiveness of the proposed framework for quantitative analysis of diffusion tensor MRI.
@Article{RSM:Mad2008,
Author = "M. Maddah and W. Grimson and S. Warfield and W. Wells",
Title = "A unified framework for clustering and quantitative
analysis of white matter fiber tracts",
Year = 2008,
Month = 04,
Abstract = "We present a novel approach for joint clustering and
point-by-point mapping of white matter fiber pathways.
Knowledge of the point correspondence along the fiber
pathways is not only necessary for accurate clustering of
the trajectories into fiber bundles, but also crucial for
any tract-oriented quantitative analysis. We employ an
expectation-maximization (EM) algorithm to cluster the
trajectories in a gamma mixture model context. The result of
clustering is the probabilistic assignment of the fiber
trajectories to each cluster, an estimate of the cluster
parameters, i.e. spatial mean and variance, and point
correspondences. The fiber bundles are modeled by the mean
trajectory and its spatial variation. Point-by-point
correspondence of the trajectories within a bundle is
obtained by constructing a distance map and a label map from
each cluster center at every iteration of the EM
algorithm. This offers a time-efficient alternative to
pairwise curve matching of all trajectories with respect to
each cluster center. The proposed method has the potential
to benefit from an anatomical atlas of fiber tracts by
incorporating it as prior information in the EM
algorithm. The algorithm is also capable of handling
outliers in a principled way. The presented results confirm
the efficiency and effectiveness of the proposed framework
for quantitative analysis of diffusion tensor MRI.",
journal = "Med Image Anal",
Pubmed = "18180197",
Keywords = "Projects:DTIModeling",
}
Downloads: 0
{"_id":"Q88GsxLByYxYboBRc","bibbaseid":"maddah-grimson-warfield-wells-aunifiedframeworkforclusteringandquantitativeanalysisofwhitematterfibertracts-2008","downloads":0,"creationDate":"2016-07-01T21:38:37.036Z","title":"A unified framework for clustering and quantitative analysis of white matter fiber tracts","author_short":["Maddah, M.","Grimson, W.","Warfield, S.","Wells, W."],"year":2008,"bibtype":"article","biburl":"http://www.sci.utah.edu/~macleod/Bibtex/biglit.bib","bibdata":{"bibtype":"article","type":"article","author":[{"firstnames":["M."],"propositions":[],"lastnames":["Maddah"],"suffixes":[]},{"firstnames":["W."],"propositions":[],"lastnames":["Grimson"],"suffixes":[]},{"firstnames":["S."],"propositions":[],"lastnames":["Warfield"],"suffixes":[]},{"firstnames":["W."],"propositions":[],"lastnames":["Wells"],"suffixes":[]}],"title":"A unified framework for clustering and quantitative analysis of white matter fiber tracts","year":"2008","month":"04","abstract":"We present a novel approach for joint clustering and point-by-point mapping of white matter fiber pathways. Knowledge of the point correspondence along the fiber pathways is not only necessary for accurate clustering of the trajectories into fiber bundles, but also crucial for any tract-oriented quantitative analysis. We employ an expectation-maximization (EM) algorithm to cluster the trajectories in a gamma mixture model context. The result of clustering is the probabilistic assignment of the fiber trajectories to each cluster, an estimate of the cluster parameters, i.e. spatial mean and variance, and point correspondences. The fiber bundles are modeled by the mean trajectory and its spatial variation. Point-by-point correspondence of the trajectories within a bundle is obtained by constructing a distance map and a label map from each cluster center at every iteration of the EM algorithm. This offers a time-efficient alternative to pairwise curve matching of all trajectories with respect to each cluster center. The proposed method has the potential to benefit from an anatomical atlas of fiber tracts by incorporating it as prior information in the EM algorithm. The algorithm is also capable of handling outliers in a principled way. The presented results confirm the efficiency and effectiveness of the proposed framework for quantitative analysis of diffusion tensor MRI.","journal":"Med Image Anal","pubmed":"18180197","keywords":"Projects:DTIModeling","bibtex":"@Article{RSM:Mad2008,\n Author\t= \"M. Maddah and W. Grimson and S. Warfield and W. Wells\",\n Title\t\t= \"A unified framework for clustering and quantitative \n\t\t analysis of white matter fiber tracts\",\n Year\t\t= 2008,\n Month\t\t= 04,\n Abstract\t= \"We present a novel approach for joint clustering and\n point-by-point mapping of white matter fiber pathways.\n\t\t Knowledge of the point correspondence along the fiber\n\t\t pathways is not only necessary for accurate clustering of\n\t\t the trajectories into fiber bundles, but also crucial for\n\t\t any tract-oriented quantitative analysis. We employ an\n\t\t expectation-maximization (EM) algorithm to cluster the\n\t\t trajectories in a gamma mixture model context. The result of\n\t\t clustering is the probabilistic assignment of the fiber\n\t\t trajectories to each cluster, an estimate of the cluster\n\t\t parameters, i.e. spatial mean and variance, and point\n\t\t correspondences. The fiber bundles are modeled by the mean\n\t\t trajectory and its spatial variation. Point-by-point\n\t\t correspondence of the trajectories within a bundle is\n\t\t obtained by constructing a distance map and a label map from\n\t\t each cluster center at every iteration of the EM\n\t\t algorithm. This offers a time-efficient alternative to\n\t\t pairwise curve matching of all trajectories with respect to\n\t\t each cluster center. The proposed method has the potential\n\t\t to benefit from an anatomical atlas of fiber tracts by\n\t\t incorporating it as prior information in the EM\n\t\t algorithm. The algorithm is also capable of handling\n\t\t outliers in a principled way. The presented results confirm\n\t\t the efficiency and effectiveness of the proposed framework\n\t\t for quantitative analysis of diffusion tensor MRI.\", \n journal\t= \"Med Image Anal\",\n Pubmed\t= \"18180197\",\n Keywords\t= \"Projects:DTIModeling\",\n}\n\n","author_short":["Maddah, M.","Grimson, W.","Warfield, S.","Wells, W."],"key":"RSM:Mad2008","id":"RSM:Mad2008","bibbaseid":"maddah-grimson-warfield-wells-aunifiedframeworkforclusteringandquantitativeanalysisofwhitematterfibertracts-2008","role":"author","urls":{},"keyword":["Projects:DTIModeling"],"downloads":0,"html":""},"search_terms":["unified","framework","clustering","quantitative","analysis","white","matter","fiber","tracts","maddah","grimson","warfield","wells"],"keywords":["projects:dtimodeling"],"authorIDs":[],"dataSources":["5HG3Kp8zRwDd7FotB"]}