Clustering of the Human Skeletal Muscle Fibers Using Linear Programming and Angular Hilbertian Metrics

Abstract : In this paper, we present a manifold clustering method for the classification of fibers obtained from diffusion tensor images (DTI) of the human skeletal muscle. Using a linear programming formulation of prototype-based clustering, we propose a novel fiber classification algo-rithm over manifolds that circumvents the necessity to embed the data in low dimensional spaces and determines automatically the number of clusters. Furthermore, we propose the use of angular Hilbertian metrics between multivariate normal distributions to define a family of distances between tensors that we generalize to fibers. These metrics are used to approximate the geodesic distances over the fiber manifold. We also discuss the case where only geodesic distances to a reduced set of landmark fibers are available. The experimental validation of the method is done using a manually annotated significant dataset of DTI of the calf muscle for healthy and diseased subjects.
Liste complète des métadonnées

https://hal-supelec.archives-ouvertes.fr/hal-00424532
Contributor : Karine El Rassi <>
Submitted on : Friday, October 16, 2009 - 3:40:42 PM
Last modification on : Tuesday, February 26, 2019 - 10:12:33 AM
Document(s) archivé(s) le : Wednesday, June 16, 2010 - 12:50:06 AM

File

Neji_ipmi09.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00424532, version 1

Collections

Citation

Radhouène Neji, Ahmed Besbes, Nikos Komodakis, Mezri Maatoouk, Jean-François Deux, et al.. Clustering of the Human Skeletal Muscle Fibers Using Linear Programming and Angular Hilbertian Metrics. 21st International Conference on Information Processing in Medical Imaging, Jul 2009, Williamsburg (Virginia), United States. pp.14-25. ⟨hal-00424532⟩

Share

Metrics

Record views

801

Files downloads

209