Prior knowledge, random walks and human skeletal muscle segmentation

Authors:
P. -Y. Baudin;N. Azzabou;P. G. Carlier;Nikos Paragios
Affiliations:
Siemens Healthcare, Saint Denis, France, Center for Vis. Comp., France, LIGM, Center for Visual Comp., Ecole des Ponts ParisTech, Univ. Paris-Est, France, Equipe Galen, INRIA Saclay, Ile-de-France ...;Institute of Myology, Paris, France, I2BM, MIRCen, IdM NMR Laboratory, CEA, Paris, France, UPMC University Paris 06, Paris, France;Institute of Myology, Paris, France, I2BM, MIRCen, IdM NMR Laboratory, CEA, Paris, France, UPMC University Paris 06, Paris, France;Center for Visual Computing, Ecole Centrale de Paris, France, LIGM (UMR CNRS), Center for Visual Computing, Ecole des Ponts ParisTech, Université Paris-Est, France, Equipe Galen, INRIA Saclay ...
Venue:
MICCAI'12 Proceedings of the 15th international conference on Medical Image Computing and Computer-Assisted Intervention - Volume Part I
Year:
2012

Citing 6
Cited 0

Multilabel Random Walker Image Segmentation Using Prior Models

CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 1 - Volume 01
Random Walks for Image Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
Prior Knowledge, Level Set Representations & Visual Grouping

International Journal of Computer Vision
Fast Musculoskeletal Registration Based on Shape Matching

MICCAI '08 Proceedings of the 11th International Conference on Medical Image Computing and Computer-Assisted Intervention, Part II
3D knowledge-based segmentation using pose-invariant higher-order graphs

MICCAI'10 Proceedings of the 13th international conference on Medical image computing and computer-assisted intervention: Part III
Probabilistic multi-shape segmentation of knee extensor and flexor muscles

MICCAI'11 Proceedings of the 14th international conference on Medical image computing and computer-assisted intervention - Volume Part III

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we propose a novel approach for segmenting the skeletal muscles in MRI automatically. In order to deal with the absence of contrast between the different muscle classes, we proposed a principled mathematical formulation that integrates prior knowledge with a random walks graph-based formulation. Prior knowledge is represented using a statistical shape atlas that once coupled with the random walks segmentation leads to an efficient iterative linear optimization system. We reveal the potential of our approach on a challenging set of real clinical data.