A novel method for 3-D segmentation and volume estimation of brain compartments from MRI

Jacob M. Agris, Rui de Figueiredo, Gilbert R. Hillman, Thomas A. Kent

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A 3-D, multiple intensity, region-growing technique is developed to segment the voxels into regions corresponding to purely white or gray matter. These voxel intensities are used to characterize the pure white and gray matter. There remains a shell of voxels between the white and gray regions. Each voxel in this shell contains a mixture of white and gray matter. A least-squares vector estimator is then used to determine the fraction of white and gray matter in each of these voxels. From these data the gray/white (g/w) ratio is calculated. An initial application of this method was able to detect g/w ratios in AIDS patients. The power of this technique lies in its ability to quantitate small changes globally affecting a volume.

Original languageEnglish (US)
Title of host publicationProceedings of the Annual Conference on Engineering in Medicine and Biology
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages68-70
Number of pages3
Editionpt 1
ISBN (Print)0780302168
StatePublished - 1991
EventProceedings of the 13th Annual International Conference of the IEEE Engineering in Medicine and Biology Society - Orlando, FL, USA
Duration: Oct 31 1991Nov 3 1991

Publication series

NameProceedings of the Annual Conference on Engineering in Medicine and Biology
Numberpt 1
Volume13
ISSN (Print)0589-1019

Other

OtherProceedings of the 13th Annual International Conference of the IEEE Engineering in Medicine and Biology Society
CityOrlando, FL, USA
Period10/31/9111/3/91

ASJC Scopus subject areas

  • Signal Processing
  • Biomedical Engineering
  • Computer Vision and Pattern Recognition
  • Health Informatics

Fingerprint

Dive into the research topics of 'A novel method for 3-D segmentation and volume estimation of brain compartments from MRI'. Together they form a unique fingerprint.

Cite this