Improving the accuracy of volumetric segmentation using pre-processing boundary detection and image reconstruction

Rick Archibald, Jiuxiang Hu, Anne Gelb, Gerald Farin

Research output: Contribution to journalArticlepeer-review

16 Scopus citations

Abstract

The concentration edge -detection and Gegenbauer image-reconstruction methods were previously shown to improve the quality of segmentation in magnetic resonance imaging. In this study, these methods are utilized as a pre-processing step to the Weibull E-SD field segmentation. It is demonstrated that the combination of the concentration edge detection and Gegenbauer reconstruction method improves the accuracy of segmentation for the simulated test data and real magnetic resonance images used in this study.

Original languageEnglish
Pages (from-to)459-466
Number of pages8
JournalIEEE Transactions on Image Processing
Volume13
Issue number4
DOIs
StatePublished - Apr 2004
Externally publishedYes

Funding

Manuscript received July 14, 2003; revised September 9, 2003. This work supported in part by the Center for System Science and Engineering Research at Arizona State University, by the Arizona Center for Alzheimer’s Disease Research, by the U.S. Defense Advanced Research Projects Agency under Grant MDA 972-00-1-0027, and by the National Science Foundation under Grant DMS01-07428, Grant EAR0222327, and Grant IIS-998016. The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Attila Kuba.

FundersFunder number
Center for System Science and Engineering Research at Arizona State University
National Science FoundationEAR0222327, IIS-998016, DMS01-07428
Defense Advanced Research Projects AgencyMDA 972-00-1-0027

    Keywords

    • Edge detection
    • Gegenbauer reconstruction
    • Magnetic resonance imaging
    • Three dimensional (3-D) segmentation
    • Weibull E-SD field

    Fingerprint

    Dive into the research topics of 'Improving the accuracy of volumetric segmentation using pre-processing boundary detection and image reconstruction'. Together they form a unique fingerprint.

    Cite this