Synthetic ground truth for validation of brain tumor MRI segmentation

Marcel Prastawa, Elizabeth Bullitt, Guido Gerig

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

    Abstract

    Validation and method of comparison for segmentation of magnetic resonance images (MRI) presenting pathology is a challenging task due to the lack of reliable ground truth. We propose a new method for generating synthetic multi-modal 3D brain MRI with tumor and edema, along with the ground truth. Tumor mass effect is modeled using a biomechanical model, while tumor and edema infiltration is modeled as a reaction-diffusion process that is guided by a modified diffusion tensor MRI. We propose the use of warping and geodesic interpolation on the diffusion tensors to simulate the displacement and the destruction of the white matter fibers. We also model the process where the contrast agent tends to accumulate in cortical csf regions and active tumor regions to obtain contrast enhanced T1w MR image that appear realistic. The result is simulated multi-modal MRI with ground truth available as sets of probability maps. The system will be able to generate large sets of simulation images with tumors of varying size, shape and location, and will additionally generate infiltrated and deformed healthy tissue probabilities.

    Original languageEnglish (US)
    Title of host publicationMedical Image Computing and Computer-Assisted Intervention - MICCAI 2005 - 8th International Conference, Proceedings
    Pages26-33
    Number of pages8
    Volume3749 LNCS
    DOIs
    StatePublished - 2005
    Event8th International Conference on Medical Image Computing and Computer-Assisted Intervention - MICCAI 2005 - Palm Springs, CA, United States
    Duration: Oct 26 2005Oct 29 2005

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume3749 LNCS
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other8th International Conference on Medical Image Computing and Computer-Assisted Intervention - MICCAI 2005
    CountryUnited States
    CityPalm Springs, CA
    Period10/26/0510/29/05

    Fingerprint

    Brain Tumor
    Magnetic Resonance Image
    Magnetic resonance
    Image segmentation
    Brain Neoplasms
    Image Segmentation
    Tumors
    Tumor
    Brain
    Magnetic Resonance Spectroscopy
    Neoplasms
    Tensors
    Edema
    Tensor
    Image Simulation
    Infiltration
    Warping
    Pathology
    Reaction-diffusion
    Accumulate

    ASJC Scopus subject areas

    • Computer Science(all)
    • Biochemistry, Genetics and Molecular Biology(all)
    • Theoretical Computer Science

    Cite this

    Prastawa, M., Bullitt, E., & Gerig, G. (2005). Synthetic ground truth for validation of brain tumor MRI segmentation. In Medical Image Computing and Computer-Assisted Intervention - MICCAI 2005 - 8th International Conference, Proceedings (Vol. 3749 LNCS, pp. 26-33). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3749 LNCS). https://doi.org/10.1007/11566465_4

    Synthetic ground truth for validation of brain tumor MRI segmentation. / Prastawa, Marcel; Bullitt, Elizabeth; Gerig, Guido.

    Medical Image Computing and Computer-Assisted Intervention - MICCAI 2005 - 8th International Conference, Proceedings. Vol. 3749 LNCS 2005. p. 26-33 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3749 LNCS).

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

    Prastawa, M, Bullitt, E & Gerig, G 2005, Synthetic ground truth for validation of brain tumor MRI segmentation. in Medical Image Computing and Computer-Assisted Intervention - MICCAI 2005 - 8th International Conference, Proceedings. vol. 3749 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 3749 LNCS, pp. 26-33, 8th International Conference on Medical Image Computing and Computer-Assisted Intervention - MICCAI 2005, Palm Springs, CA, United States, 10/26/05. https://doi.org/10.1007/11566465_4
    Prastawa M, Bullitt E, Gerig G. Synthetic ground truth for validation of brain tumor MRI segmentation. In Medical Image Computing and Computer-Assisted Intervention - MICCAI 2005 - 8th International Conference, Proceedings. Vol. 3749 LNCS. 2005. p. 26-33. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/11566465_4
    Prastawa, Marcel ; Bullitt, Elizabeth ; Gerig, Guido. / Synthetic ground truth for validation of brain tumor MRI segmentation. Medical Image Computing and Computer-Assisted Intervention - MICCAI 2005 - 8th International Conference, Proceedings. Vol. 3749 LNCS 2005. pp. 26-33 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{c6214f61516d4922996165ef471f03bb,
    title = "Synthetic ground truth for validation of brain tumor MRI segmentation",
    abstract = "Validation and method of comparison for segmentation of magnetic resonance images (MRI) presenting pathology is a challenging task due to the lack of reliable ground truth. We propose a new method for generating synthetic multi-modal 3D brain MRI with tumor and edema, along with the ground truth. Tumor mass effect is modeled using a biomechanical model, while tumor and edema infiltration is modeled as a reaction-diffusion process that is guided by a modified diffusion tensor MRI. We propose the use of warping and geodesic interpolation on the diffusion tensors to simulate the displacement and the destruction of the white matter fibers. We also model the process where the contrast agent tends to accumulate in cortical csf regions and active tumor regions to obtain contrast enhanced T1w MR image that appear realistic. The result is simulated multi-modal MRI with ground truth available as sets of probability maps. The system will be able to generate large sets of simulation images with tumors of varying size, shape and location, and will additionally generate infiltrated and deformed healthy tissue probabilities.",
    author = "Marcel Prastawa and Elizabeth Bullitt and Guido Gerig",
    year = "2005",
    doi = "10.1007/11566465_4",
    language = "English (US)",
    isbn = "3540293272",
    volume = "3749 LNCS",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    pages = "26--33",
    booktitle = "Medical Image Computing and Computer-Assisted Intervention - MICCAI 2005 - 8th International Conference, Proceedings",

    }

    TY - GEN

    T1 - Synthetic ground truth for validation of brain tumor MRI segmentation

    AU - Prastawa, Marcel

    AU - Bullitt, Elizabeth

    AU - Gerig, Guido

    PY - 2005

    Y1 - 2005

    N2 - Validation and method of comparison for segmentation of magnetic resonance images (MRI) presenting pathology is a challenging task due to the lack of reliable ground truth. We propose a new method for generating synthetic multi-modal 3D brain MRI with tumor and edema, along with the ground truth. Tumor mass effect is modeled using a biomechanical model, while tumor and edema infiltration is modeled as a reaction-diffusion process that is guided by a modified diffusion tensor MRI. We propose the use of warping and geodesic interpolation on the diffusion tensors to simulate the displacement and the destruction of the white matter fibers. We also model the process where the contrast agent tends to accumulate in cortical csf regions and active tumor regions to obtain contrast enhanced T1w MR image that appear realistic. The result is simulated multi-modal MRI with ground truth available as sets of probability maps. The system will be able to generate large sets of simulation images with tumors of varying size, shape and location, and will additionally generate infiltrated and deformed healthy tissue probabilities.

    AB - Validation and method of comparison for segmentation of magnetic resonance images (MRI) presenting pathology is a challenging task due to the lack of reliable ground truth. We propose a new method for generating synthetic multi-modal 3D brain MRI with tumor and edema, along with the ground truth. Tumor mass effect is modeled using a biomechanical model, while tumor and edema infiltration is modeled as a reaction-diffusion process that is guided by a modified diffusion tensor MRI. We propose the use of warping and geodesic interpolation on the diffusion tensors to simulate the displacement and the destruction of the white matter fibers. We also model the process where the contrast agent tends to accumulate in cortical csf regions and active tumor regions to obtain contrast enhanced T1w MR image that appear realistic. The result is simulated multi-modal MRI with ground truth available as sets of probability maps. The system will be able to generate large sets of simulation images with tumors of varying size, shape and location, and will additionally generate infiltrated and deformed healthy tissue probabilities.

    UR - http://www.scopus.com/inward/record.url?scp=33744825479&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=33744825479&partnerID=8YFLogxK

    U2 - 10.1007/11566465_4

    DO - 10.1007/11566465_4

    M3 - Conference contribution

    SN - 3540293272

    SN - 9783540293279

    VL - 3749 LNCS

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

    SP - 26

    EP - 33

    BT - Medical Image Computing and Computer-Assisted Intervention - MICCAI 2005 - 8th International Conference, Proceedings

    ER -