Model-based brain and tumor segmentation

Nathan Moon, Elizabeth Bullitt, Koen Van Leemput, Guido Gerig

    Research output: Chapter in Book/Report/Conference proceedingChapter

    Abstract

    Combining image segmentation based on statistical classification with a geometric prior has been shown to significantly increase robustness and reproducibility. Using a probabilistic geometric model and image registration serves both initialization of probability density functions and definition of spatial constraints. A strong spatial prior, however, prevents segmentation of structures that are not part of the model. Our driving application is the segmentation of brain tissue and tumors from three-dimensional magnetic resonance imaging (MRI). Our goal is a high-quality segmentation of both healthy tissue and tumor. We present an extension to an existing expectation maximization (EM) segmentation algorithm that modifies a probabilistic brain atlas with an individual subject's information about tumor location obtained from subtraction of post- and pre-contrast MRI. The new method handles various types of pathology, space-occupying mass tumors and infiltrating changes like edema. Preliminary results on five cases presenting tumor types with very different characteristics demonstrate the potential of the new technique for clinical routine use for planning and monitoring in neurosurgery, radiation oncology, and radiology.

    Original languageEnglish (US)
    Title of host publicationProceedings - International Conference on Pattern Recognition
    Pages528-531
    Number of pages4
    Volume16
    Edition1
    StatePublished - 2002

    Fingerprint

    Tumors
    Brain
    Magnetic resonance
    Tissue
    Neurosurgery
    Imaging techniques
    Radiology
    Oncology
    Image registration
    Pathology
    Image segmentation
    Probability density function
    Radiation
    Planning
    Monitoring

    ASJC Scopus subject areas

    • Electrical and Electronic Engineering
    • Computer Vision and Pattern Recognition
    • Hardware and Architecture

    Cite this

    Moon, N., Bullitt, E., Van Leemput, K., & Gerig, G. (2002). Model-based brain and tumor segmentation. In Proceedings - International Conference on Pattern Recognition (1 ed., Vol. 16, pp. 528-531)

    Model-based brain and tumor segmentation. / Moon, Nathan; Bullitt, Elizabeth; Van Leemput, Koen; Gerig, Guido.

    Proceedings - International Conference on Pattern Recognition. Vol. 16 1. ed. 2002. p. 528-531.

    Research output: Chapter in Book/Report/Conference proceedingChapter

    Moon, N, Bullitt, E, Van Leemput, K & Gerig, G 2002, Model-based brain and tumor segmentation. in Proceedings - International Conference on Pattern Recognition. 1 edn, vol. 16, pp. 528-531.
    Moon N, Bullitt E, Van Leemput K, Gerig G. Model-based brain and tumor segmentation. In Proceedings - International Conference on Pattern Recognition. 1 ed. Vol. 16. 2002. p. 528-531
    Moon, Nathan ; Bullitt, Elizabeth ; Van Leemput, Koen ; Gerig, Guido. / Model-based brain and tumor segmentation. Proceedings - International Conference on Pattern Recognition. Vol. 16 1. ed. 2002. pp. 528-531
    @inbook{3493548a86e546179c45764ef5e652ec,
    title = "Model-based brain and tumor segmentation",
    abstract = "Combining image segmentation based on statistical classification with a geometric prior has been shown to significantly increase robustness and reproducibility. Using a probabilistic geometric model and image registration serves both initialization of probability density functions and definition of spatial constraints. A strong spatial prior, however, prevents segmentation of structures that are not part of the model. Our driving application is the segmentation of brain tissue and tumors from three-dimensional magnetic resonance imaging (MRI). Our goal is a high-quality segmentation of both healthy tissue and tumor. We present an extension to an existing expectation maximization (EM) segmentation algorithm that modifies a probabilistic brain atlas with an individual subject's information about tumor location obtained from subtraction of post- and pre-contrast MRI. The new method handles various types of pathology, space-occupying mass tumors and infiltrating changes like edema. Preliminary results on five cases presenting tumor types with very different characteristics demonstrate the potential of the new technique for clinical routine use for planning and monitoring in neurosurgery, radiation oncology, and radiology.",
    author = "Nathan Moon and Elizabeth Bullitt and {Van Leemput}, Koen and Guido Gerig",
    year = "2002",
    language = "English (US)",
    volume = "16",
    pages = "528--531",
    booktitle = "Proceedings - International Conference on Pattern Recognition",
    edition = "1",

    }

    TY - CHAP

    T1 - Model-based brain and tumor segmentation

    AU - Moon, Nathan

    AU - Bullitt, Elizabeth

    AU - Van Leemput, Koen

    AU - Gerig, Guido

    PY - 2002

    Y1 - 2002

    N2 - Combining image segmentation based on statistical classification with a geometric prior has been shown to significantly increase robustness and reproducibility. Using a probabilistic geometric model and image registration serves both initialization of probability density functions and definition of spatial constraints. A strong spatial prior, however, prevents segmentation of structures that are not part of the model. Our driving application is the segmentation of brain tissue and tumors from three-dimensional magnetic resonance imaging (MRI). Our goal is a high-quality segmentation of both healthy tissue and tumor. We present an extension to an existing expectation maximization (EM) segmentation algorithm that modifies a probabilistic brain atlas with an individual subject's information about tumor location obtained from subtraction of post- and pre-contrast MRI. The new method handles various types of pathology, space-occupying mass tumors and infiltrating changes like edema. Preliminary results on five cases presenting tumor types with very different characteristics demonstrate the potential of the new technique for clinical routine use for planning and monitoring in neurosurgery, radiation oncology, and radiology.

    AB - Combining image segmentation based on statistical classification with a geometric prior has been shown to significantly increase robustness and reproducibility. Using a probabilistic geometric model and image registration serves both initialization of probability density functions and definition of spatial constraints. A strong spatial prior, however, prevents segmentation of structures that are not part of the model. Our driving application is the segmentation of brain tissue and tumors from three-dimensional magnetic resonance imaging (MRI). Our goal is a high-quality segmentation of both healthy tissue and tumor. We present an extension to an existing expectation maximization (EM) segmentation algorithm that modifies a probabilistic brain atlas with an individual subject's information about tumor location obtained from subtraction of post- and pre-contrast MRI. The new method handles various types of pathology, space-occupying mass tumors and infiltrating changes like edema. Preliminary results on five cases presenting tumor types with very different characteristics demonstrate the potential of the new technique for clinical routine use for planning and monitoring in neurosurgery, radiation oncology, and radiology.

    UR - http://www.scopus.com/inward/record.url?scp=33751574210&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=33751574210&partnerID=8YFLogxK

    M3 - Chapter

    AN - SCOPUS:33751574210

    VL - 16

    SP - 528

    EP - 531

    BT - Proceedings - International Conference on Pattern Recognition

    ER -