Laidlaw, David H. and Fleischer, Kurt W. and Barr, Alan H. (1997) Partial-Volume Bayesian Classification of Material Mixtures in MR Volume Data using Voxel Histograms. California Institute of Technology , Pasadena, CA. (Unpublished) http://resolver.caltech.edu/CaltechCSTR:1997.cs-tr-97-12
- Submitted Version
See Usage Policy.
- Submitted Version
See Usage Policy.
Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechCSTR:1997.cs-tr-97-12
We present a new algorithm for identifying the distribution of different material types in volumetric datasets such as those produced with Magnetic Resonance Imaging (MRI) or Computed Tomography (CT). Because we allow for mixtures of materials and treat voxels as regions, our technique reduces the classification artifacts that thresholding can create along boundaries between materials and is particularly useful for creating accurate geometric models and renderings from volume data. It also has the potential to make more-accurate volume measurements and classifies noisy, low-resolution data well. There are two unusual aspects to our approach. First, we assume that, due to partial-volume effects, voxels can contain more than one material, e.g., both muscle and fat; we compute the relative proportion of each material in the voxels. Second, we incorporate information from neighboring voxels into the classification process by reconstructing a continuous function, p(x), from the samples and then looking at the distribution of values that p takes on within the region of a voxel. This distribution of values is represented by a histogram taken over the region of the voxel; the mixture of materials that those values measure is identified within the voxel using a probabilistic Bayesian approach that matches the histogram by finding the mixture of materials within each voxel most likely to have created the histogram. The size of regions that we classify is chosen to match the spacing of the samples because the spacing is intrinsically related to the minimum feature size that the reconstructed continuous function can represent.
|Item Type:||Report or Paper (Technical Report)|
|Additional Information:||© 1997 California Institute of Technology. Many thanks to Matthew Avalos, who has been instrumental in implementation. Thanks to Barbara Meier, David Kirk, John Snyder, Bena Currin, and Mark Montague for reviewing early drafts and making suggestions. Thanks also to Allen Corcoran, Constance Villani, Cindy Ball, and Eric Winfree for production help, and Jose Jimenez for the late-night MR sessions. Our data was collected in collaboration with the Huntington Magnetic Resonance Center and the Caltech Biological Imaging Center, both in Pasadena. This work was supported in part by grants from Apple, DEC, Hewlett Packard, and IBM. Additional support was provided by NSF (ASC-89-20219) as part of the NSF/ARPA STC for Computer Graphics and Scientific Visualization, by the DOE (DE-FG03-92ER25134) as part of the Center for Research in Computational Biology, by the National Institute on Drug Abuse and the National Institute of Mental Health as part of the Human Brain Project, and by the Beckman Institute Foundation. All opinions, findings, conclusions, or recommendations expressed in this document are those of the authors and do not necessarily reflect the views of the sponsoring agencies.|
|Group:||Computer Science Technical Reports|
|Usage Policy:||You are granted permission for individual, educational, research and non-commercial reproduction, distribution, display and performance of this work in any format.|
|Deposited By:||Imported from CaltechCSTR|
|Deposited On:||25 Apr 2001|
|Last Modified:||27 Mar 2017 19:10|
Repository Staff Only: item control page