Automated Brain-Tissue Segmentation by Multi-Feature SVM Classification
logo

Please use this identifier to cite or link to this publication: http://hdl.handle.net/10380/3443
We present a method for automated brain-tissue segmentation through voxelwise classification. Our algorithm uses manually labeled training images to train a support vector machine (SVM) classifier, which is then used for the segmentation of target images. The classification incorporates voxel intensities from a T1-weighted scan, an IR scan, and a FLAIR scan; features to encode the voxel position in the image; and Gaussian-scale-space features and Gaussian-derivative features at
multiple scales to facilitate a smooth segmentation.
An experiment on data from the MRBrainS13 brain-tissue-segmentation challenge showed that our algorithm produces reasonable segmentations in a reasonable amount of time.

Reviews
There is no review at this time. Be the first to review this publication!

Quick Comments


Resources
backyellow
Download All
Download Paper , View Paper

Statistics more
backyellow
Global rating: starstarstarstarstar
Review rating: starstarstarstarstar [review]
Paper Quality: plus minus

Information more
backyellow
Categories: Classification, Mathematics, Segmentation
Keywords: Machine Learning, Pattern Recognition, Classification, Segmentation, Brain, MRI
Tracking Number: NWO 639.022.010
Export citation:

Share
backyellow
Share

Linked Publications more
backyellow
The graph windowed Fourier transform: a tool to quantify the gyrification of the cerebral cortex The graph windowed Fourier transform: a tool to quantify the gyrification of the cerebral cortex
by Rabiei H., Richard F., Roth M., Anton J., Coulon O., Lefevre J.
Tensor Spectral Matching of Diffusion Weighted Images Tensor Spectral Matching of Diffusion Weighted Images
by Orasanu E., Melbourne A., Lorenzi M., Modat M., Lombaert H., Eaton-Rosen Z., Robertson N.J., Kendall G.S., Marlow N., Ourselin S.

View license
Loading license...

Send a message to the author
main_flat
Powered by Midas