Automatic brain and tumor segmentation

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

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

    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 of sought structures 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. In practical applications, we encounter either the presentation of new objects that cannot be modeled with a spatial prior or regional intensity changes of existing structures not explained by 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 healthy tissue and a precise delineation of tumor boundaries. 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 publicationMedical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings
    PublisherSpringer Verlag
    Pages372-379
    Number of pages8
    Volume2488
    ISBN (Print)9783540457862
    StatePublished - 2002
    Event5th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2002 - Tokyo, Japan
    Duration: Sep 25 2002Sep 28 2002

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume2488
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other5th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2002
    CountryJapan
    CityTokyo
    Period9/25/029/28/02

    Fingerprint

    Tumors
    Tumor
    Brain
    Segmentation
    Magnetic Resonance Imaging
    Magnetic resonance
    Tissue
    Neurosurgery
    Three-dimensional Imaging
    Imaging techniques
    Oncology
    Radiology
    Expectation Maximization
    Geometric Model
    Atlas
    Image registration
    Reproducibility
    Image Registration
    Pathology
    Subtraction

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Moon, N., Bullitt, E., van Leemput, K., & Gerig, G. (2002). Automatic brain and tumor segmentation. In Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings (Vol. 2488, pp. 372-379). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2488). Springer Verlag.

    Automatic brain and tumor segmentation. / Moon, Nathan; Bullitt, Elizabeth; van Leemput, Koen; Gerig, Guido.

    Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings. Vol. 2488 Springer Verlag, 2002. p. 372-379 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2488).

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

    Moon, N, Bullitt, E, van Leemput, K & Gerig, G 2002, Automatic brain and tumor segmentation. in Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings. vol. 2488, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2488, Springer Verlag, pp. 372-379, 5th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2002, Tokyo, Japan, 9/25/02.
    Moon N, Bullitt E, van Leemput K, Gerig G. Automatic brain and tumor segmentation. In Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings. Vol. 2488. Springer Verlag. 2002. p. 372-379. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    Moon, Nathan ; Bullitt, Elizabeth ; van Leemput, Koen ; Gerig, Guido. / Automatic brain and tumor segmentation. Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings. Vol. 2488 Springer Verlag, 2002. pp. 372-379 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{53d95228819041b3b506d3d496d4754f,
    title = "Automatic 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 of sought structures 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. In practical applications, we encounter either the presentation of new objects that cannot be modeled with a spatial prior or regional intensity changes of existing structures not explained by 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 healthy tissue and a precise delineation of tumor boundaries. 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)",
    isbn = "9783540457862",
    volume = "2488",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    publisher = "Springer Verlag",
    pages = "372--379",
    booktitle = "Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings",
    address = "Germany",

    }

    TY - GEN

    T1 - Automatic 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 of sought structures 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. In practical applications, we encounter either the presentation of new objects that cannot be modeled with a spatial prior or regional intensity changes of existing structures not explained by 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 healthy tissue and a precise delineation of tumor boundaries. 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 of sought structures 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. In practical applications, we encounter either the presentation of new objects that cannot be modeled with a spatial prior or regional intensity changes of existing structures not explained by 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 healthy tissue and a precise delineation of tumor boundaries. 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=84974717451&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84974717451&partnerID=8YFLogxK

    M3 - Conference contribution

    AN - SCOPUS:84974717451

    SN - 9783540457862

    VL - 2488

    T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    SP - 372

    EP - 379

    BT - Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings

    PB - Springer Verlag

    ER -