3D tensor normalization for improved accuracy in DTI tensor registration methods

Aditya Gupta, Maria Escolar, Cheryl Dietrich, John Gilmore, Guido Gerig, Martin Styner

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

    Abstract

    This paper presents a method for normalization of diffusion tensor images (DTI) to a fixed DTI template, a pre-processing step to improve the performance of full tensor based registration methods. The proposed method maps the individual tensors of the subject image in to the template space based on matching the cumulative distribution function and the fractional anisotrophy values. The method aims to determine a more accurate deformation field from any full tensor registration method by applying the registration algorithm on the normalized DTI rather than the original DTI. The deformation field applied to the original tensor images are compared to the deformed image without normalization for 11 different cases of mapping seven subjects (neonate through 2 years) to two different atlases. The method shows an improvement in DTI registration based on comparing the normalized fractional anisotropy values of major fiber tracts in the brain.

    Original languageEnglish (US)
    Title of host publicationBiomedical Image Registration - 5th International Workshop, WBIR 2012, Proceedings
    Pages170-179
    Number of pages10
    Volume7359 LNCS
    DOIs
    StatePublished - 2012
    Event5th International Workshop on Biomedical Image Registration, WBIR 2012 - Nashville, TN, United States
    Duration: Jul 7 2012Jul 8 2012

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume7359 LNCS
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other5th International Workshop on Biomedical Image Registration, WBIR 2012
    CountryUnited States
    CityNashville, TN
    Period7/7/127/8/12

    Fingerprint

    Registration
    Normalization
    Tensors
    Tensor
    Template
    Fractional
    Atlas
    Image registration
    Cumulative distribution function
    Image Registration
    Distribution functions
    Preprocessing
    Anisotropy
    Brain
    Fiber
    Fibers
    Processing

    Keywords

    • DTI Registration
    • DTITK
    • Tensor Normalization

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Gupta, A., Escolar, M., Dietrich, C., Gilmore, J., Gerig, G., & Styner, M. (2012). 3D tensor normalization for improved accuracy in DTI tensor registration methods. In Biomedical Image Registration - 5th International Workshop, WBIR 2012, Proceedings (Vol. 7359 LNCS, pp. 170-179). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7359 LNCS). https://doi.org/10.1007/978-3-642-31340-0_18

    3D tensor normalization for improved accuracy in DTI tensor registration methods. / Gupta, Aditya; Escolar, Maria; Dietrich, Cheryl; Gilmore, John; Gerig, Guido; Styner, Martin.

    Biomedical Image Registration - 5th International Workshop, WBIR 2012, Proceedings. Vol. 7359 LNCS 2012. p. 170-179 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7359 LNCS).

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

    Gupta, A, Escolar, M, Dietrich, C, Gilmore, J, Gerig, G & Styner, M 2012, 3D tensor normalization for improved accuracy in DTI tensor registration methods. in Biomedical Image Registration - 5th International Workshop, WBIR 2012, Proceedings. vol. 7359 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7359 LNCS, pp. 170-179, 5th International Workshop on Biomedical Image Registration, WBIR 2012, Nashville, TN, United States, 7/7/12. https://doi.org/10.1007/978-3-642-31340-0_18
    Gupta A, Escolar M, Dietrich C, Gilmore J, Gerig G, Styner M. 3D tensor normalization for improved accuracy in DTI tensor registration methods. In Biomedical Image Registration - 5th International Workshop, WBIR 2012, Proceedings. Vol. 7359 LNCS. 2012. p. 170-179. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-31340-0_18
    Gupta, Aditya ; Escolar, Maria ; Dietrich, Cheryl ; Gilmore, John ; Gerig, Guido ; Styner, Martin. / 3D tensor normalization for improved accuracy in DTI tensor registration methods. Biomedical Image Registration - 5th International Workshop, WBIR 2012, Proceedings. Vol. 7359 LNCS 2012. pp. 170-179 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{014b41cb2dae40dbbe69af67f4bcb487,
    title = "3D tensor normalization for improved accuracy in DTI tensor registration methods",
    abstract = "This paper presents a method for normalization of diffusion tensor images (DTI) to a fixed DTI template, a pre-processing step to improve the performance of full tensor based registration methods. The proposed method maps the individual tensors of the subject image in to the template space based on matching the cumulative distribution function and the fractional anisotrophy values. The method aims to determine a more accurate deformation field from any full tensor registration method by applying the registration algorithm on the normalized DTI rather than the original DTI. The deformation field applied to the original tensor images are compared to the deformed image without normalization for 11 different cases of mapping seven subjects (neonate through 2 years) to two different atlases. The method shows an improvement in DTI registration based on comparing the normalized fractional anisotropy values of major fiber tracts in the brain.",
    keywords = "DTI Registration, DTITK, Tensor Normalization",
    author = "Aditya Gupta and Maria Escolar and Cheryl Dietrich and John Gilmore and Guido Gerig and Martin Styner",
    year = "2012",
    doi = "10.1007/978-3-642-31340-0_18",
    language = "English (US)",
    isbn = "9783642313394",
    volume = "7359 LNCS",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    pages = "170--179",
    booktitle = "Biomedical Image Registration - 5th International Workshop, WBIR 2012, Proceedings",

    }

    TY - GEN

    T1 - 3D tensor normalization for improved accuracy in DTI tensor registration methods

    AU - Gupta, Aditya

    AU - Escolar, Maria

    AU - Dietrich, Cheryl

    AU - Gilmore, John

    AU - Gerig, Guido

    AU - Styner, Martin

    PY - 2012

    Y1 - 2012

    N2 - This paper presents a method for normalization of diffusion tensor images (DTI) to a fixed DTI template, a pre-processing step to improve the performance of full tensor based registration methods. The proposed method maps the individual tensors of the subject image in to the template space based on matching the cumulative distribution function and the fractional anisotrophy values. The method aims to determine a more accurate deformation field from any full tensor registration method by applying the registration algorithm on the normalized DTI rather than the original DTI. The deformation field applied to the original tensor images are compared to the deformed image without normalization for 11 different cases of mapping seven subjects (neonate through 2 years) to two different atlases. The method shows an improvement in DTI registration based on comparing the normalized fractional anisotropy values of major fiber tracts in the brain.

    AB - This paper presents a method for normalization of diffusion tensor images (DTI) to a fixed DTI template, a pre-processing step to improve the performance of full tensor based registration methods. The proposed method maps the individual tensors of the subject image in to the template space based on matching the cumulative distribution function and the fractional anisotrophy values. The method aims to determine a more accurate deformation field from any full tensor registration method by applying the registration algorithm on the normalized DTI rather than the original DTI. The deformation field applied to the original tensor images are compared to the deformed image without normalization for 11 different cases of mapping seven subjects (neonate through 2 years) to two different atlases. The method shows an improvement in DTI registration based on comparing the normalized fractional anisotropy values of major fiber tracts in the brain.

    KW - DTI Registration

    KW - DTITK

    KW - Tensor Normalization

    UR - http://www.scopus.com/inward/record.url?scp=84863997773&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84863997773&partnerID=8YFLogxK

    U2 - 10.1007/978-3-642-31340-0_18

    DO - 10.1007/978-3-642-31340-0_18

    M3 - Conference contribution

    SN - 9783642313394

    VL - 7359 LNCS

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

    SP - 170

    EP - 179

    BT - Biomedical Image Registration - 5th International Workshop, WBIR 2012, Proceedings

    ER -