System for MR brain image segmentation

Atam P. Dhawan, Aleksandar Zavaljevski, Alok Sarwal, Scott Holland, Mary Gaskill-Shipley, William S. Ball

Research output: Contribution to journalConference articlepeer-review

6 Scopus citations


A system for manual segmentation of multispectral MR images into anatomical structures is described. The system consists of a SPARC workstation and a PIXAR image computer. The system simultaneously displays an enlarged working area as well as T1 weighted, and short and long echo T2 weighted images. Multiple linked cursors are also displayed for easier orientation. The operator can easily position the working area, specify the region in the working area, assign a label to the chosen region, and save the results in a file. The sophisticated man-machine interface, along with flexible choice of displayed images of the brain and their resolutions, make the system very precise and productive for manual segmentation with the interobserver variability less than 5%.

Original languageEnglish (US)
Pages (from-to)732-733
Number of pages2
JournalAnnual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings
StatePublished - 1996
Externally publishedYes
EventProceedings of the 1996 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Part 2 (of 5) - Amsterdam, Neth
Duration: Oct 31 1996Nov 3 1996

All Science Journal Classification (ASJC) codes

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


Dive into the research topics of 'System for MR brain image segmentation'. Together they form a unique fingerprint.

Cite this