MR image segmentation using phase information and a novel multiscale scheme

Med Image Comput Comput Assist Interv. 2006;9(Pt 2):920-7. doi: 10.1007/11866763_113.

Abstract

This paper considers the problem of automatic classification of textured tissues in 3D MRI. More specifically, it aims at validating the use of features extracted from the phase of the MR signal to improve texture discrimination in bone segmentation. This extra information provides better segmentation, compared to using magnitude only features. We also present a novel multiscale scheme to improve the speed of pixel based classification algorithm, such as support vector machines. This algorithm dramatically increases the speed of the segmentation process by an order of magnitude through a reduction of the number of pixels that needs to be classified in the image.

Publication types

  • Evaluation Study

MeSH terms

  • Algorithms*
  • Artificial Intelligence*
  • Humans
  • Image Enhancement / methods
  • Image Interpretation, Computer-Assisted / methods*
  • Imaging, Three-Dimensional / methods*
  • Knee Joint / anatomy & histology*
  • Magnetic Resonance Imaging / methods*
  • Pattern Recognition, Automated / methods*
  • Reproducibility of Results
  • Sensitivity and Specificity