Prostate MR image segmentation using 3D active appearance models

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    376 Downloads (Pure)

    Abstract

    This paper presents a method for automatic segmentation of the prostate from transversal T2-weighted images based on 3D Active Appearance Models (AAM). The algorithm consist of two stages. Firstly, Shape Context based non-rigid surface registration of the manual segmented images is used to obtain the point correspondence between the given training cases. Subsequently, an AAM is used to segment the prostate on 50 training cases. The method is evaluated using a 5-fold cross validation over 5 repetitions. The mean Dice similarity coefficient and 95% Hausdorff distance are 0.78 and 7.32 mm respectively.
    Original languageUndefined
    Title of host publicationPROMISE 12 : Miccai 2012 Grand Challenge onProstate MR Image SEgmentation
    PublisherPROMISE12
    Pages44-51
    Number of pages8
    ISBN (Print)not assigned
    Publication statusPublished - 1 Oct 2012
    EventPRostate MR Image SEgmentation, PROMISE 2012 - Nice, France
    Duration: 1 Oct 20121 Oct 2012

    Publication series

    NameGrand Challenges in Medical Image Analysis
    PublisherPROMISE12

    Workshop

    WorkshopPRostate MR Image SEgmentation, PROMISE 2012
    Period1/10/121/10/12
    OtherOctober 1, 2012

    Keywords

    • METIS-289733
    • Active Appearance Model
    • EWI-22366
    • Shape Context
    • Prostate segmentation
    • IR-83366

    Cite this