Brain lesion segmentation through physical model estimation

Marcel Prastawa, Guido Gerig

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

    Abstract

    Segmentations of brain lesions from Magnetic Resonance (MR) images is crucial for quantitative analysis of lesion populations in neuroimaging of neurological disorders. We propose a new method for segmenting lesions in brain MRI by inferring the underlying physical models for pathology. We use the reaction-diffusion model as our physical model, where the diffusion process is guided by real diffusion tensor fields that are obtained from Diffusion Tensor Imaging (DTI). The method perorms segmentation by solving the inverse problem, where it determines the optimal parameters for the physical model that generates the observed image. We show that the proposed method can infer reasonable models for multiple sclerosis (MS) lesions and healthy MRI data. The method has potential for further extensions with different physical models or even non-physical models based on existing segmentation schemes.

    Original languageEnglish (US)
    Title of host publicationAdvances in Visual Computing - 4th International Symposium, ISVC 2008, Proceedings
    Pages562-571
    Number of pages10
    Volume5358 LNCS
    EditionPART 1
    DOIs
    StatePublished - 2008
    Event4th International Symposium on Visual Computing, ISVC 2008 - Las Vegas, NV, United States
    Duration: Dec 1 2008Dec 3 2008

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    NumberPART 1
    Volume5358 LNCS
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other4th International Symposium on Visual Computing, ISVC 2008
    CountryUnited States
    CityLas Vegas, NV
    Period12/1/0812/3/08

    Fingerprint

    Physical Model
    Brain
    Segmentation
    Tensor
    Multiple Sclerosis
    Neuroimaging
    Magnetic Resonance Image
    Reaction-diffusion Model
    Magnetic resonance imaging
    Optimal Parameter
    Quantitative Analysis
    Diffusion Process
    Disorder
    Inverse Problem
    Diffusion tensor imaging
    Imaging
    Model-based
    Pathology
    Magnetic resonance
    Inverse problems

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Prastawa, M., & Gerig, G. (2008). Brain lesion segmentation through physical model estimation. In Advances in Visual Computing - 4th International Symposium, ISVC 2008, Proceedings (PART 1 ed., Vol. 5358 LNCS, pp. 562-571). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5358 LNCS, No. PART 1). https://doi.org/10.1007/978-3-540-89639-5_54

    Brain lesion segmentation through physical model estimation. / Prastawa, Marcel; Gerig, Guido.

    Advances in Visual Computing - 4th International Symposium, ISVC 2008, Proceedings. Vol. 5358 LNCS PART 1. ed. 2008. p. 562-571 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5358 LNCS, No. PART 1).

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

    Prastawa, M & Gerig, G 2008, Brain lesion segmentation through physical model estimation. in Advances in Visual Computing - 4th International Symposium, ISVC 2008, Proceedings. PART 1 edn, vol. 5358 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 1, vol. 5358 LNCS, pp. 562-571, 4th International Symposium on Visual Computing, ISVC 2008, Las Vegas, NV, United States, 12/1/08. https://doi.org/10.1007/978-3-540-89639-5_54
    Prastawa M, Gerig G. Brain lesion segmentation through physical model estimation. In Advances in Visual Computing - 4th International Symposium, ISVC 2008, Proceedings. PART 1 ed. Vol. 5358 LNCS. 2008. p. 562-571. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). https://doi.org/10.1007/978-3-540-89639-5_54
    Prastawa, Marcel ; Gerig, Guido. / Brain lesion segmentation through physical model estimation. Advances in Visual Computing - 4th International Symposium, ISVC 2008, Proceedings. Vol. 5358 LNCS PART 1. ed. 2008. pp. 562-571 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1).
    @inproceedings{83c4b20bd2814abfb77431c00a455f46,
    title = "Brain lesion segmentation through physical model estimation",
    abstract = "Segmentations of brain lesions from Magnetic Resonance (MR) images is crucial for quantitative analysis of lesion populations in neuroimaging of neurological disorders. We propose a new method for segmenting lesions in brain MRI by inferring the underlying physical models for pathology. We use the reaction-diffusion model as our physical model, where the diffusion process is guided by real diffusion tensor fields that are obtained from Diffusion Tensor Imaging (DTI). The method perorms segmentation by solving the inverse problem, where it determines the optimal parameters for the physical model that generates the observed image. We show that the proposed method can infer reasonable models for multiple sclerosis (MS) lesions and healthy MRI data. The method has potential for further extensions with different physical models or even non-physical models based on existing segmentation schemes.",
    author = "Marcel Prastawa and Guido Gerig",
    year = "2008",
    doi = "10.1007/978-3-540-89639-5_54",
    language = "English (US)",
    isbn = "3540896384",
    volume = "5358 LNCS",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    number = "PART 1",
    pages = "562--571",
    booktitle = "Advances in Visual Computing - 4th International Symposium, ISVC 2008, Proceedings",
    edition = "PART 1",

    }

    TY - GEN

    T1 - Brain lesion segmentation through physical model estimation

    AU - Prastawa, Marcel

    AU - Gerig, Guido

    PY - 2008

    Y1 - 2008

    N2 - Segmentations of brain lesions from Magnetic Resonance (MR) images is crucial for quantitative analysis of lesion populations in neuroimaging of neurological disorders. We propose a new method for segmenting lesions in brain MRI by inferring the underlying physical models for pathology. We use the reaction-diffusion model as our physical model, where the diffusion process is guided by real diffusion tensor fields that are obtained from Diffusion Tensor Imaging (DTI). The method perorms segmentation by solving the inverse problem, where it determines the optimal parameters for the physical model that generates the observed image. We show that the proposed method can infer reasonable models for multiple sclerosis (MS) lesions and healthy MRI data. The method has potential for further extensions with different physical models or even non-physical models based on existing segmentation schemes.

    AB - Segmentations of brain lesions from Magnetic Resonance (MR) images is crucial for quantitative analysis of lesion populations in neuroimaging of neurological disorders. We propose a new method for segmenting lesions in brain MRI by inferring the underlying physical models for pathology. We use the reaction-diffusion model as our physical model, where the diffusion process is guided by real diffusion tensor fields that are obtained from Diffusion Tensor Imaging (DTI). The method perorms segmentation by solving the inverse problem, where it determines the optimal parameters for the physical model that generates the observed image. We show that the proposed method can infer reasonable models for multiple sclerosis (MS) lesions and healthy MRI data. The method has potential for further extensions with different physical models or even non-physical models based on existing segmentation schemes.

    UR - http://www.scopus.com/inward/record.url?scp=68649092290&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=68649092290&partnerID=8YFLogxK

    U2 - 10.1007/978-3-540-89639-5_54

    DO - 10.1007/978-3-540-89639-5_54

    M3 - Conference contribution

    AN - SCOPUS:68649092290

    SN - 3540896384

    SN - 9783540896388

    VL - 5358 LNCS

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

    SP - 562

    EP - 571

    BT - Advances in Visual Computing - 4th International Symposium, ISVC 2008, Proceedings

    ER -