FADTTS: Functional analysis of diffusion tensor tract statistics

Hongtu Zhu, Linglong Kong, Runze Li, Martin Styner, Guido Gerig, Weili Lin, John H. Gilmore

    Research output: Contribution to journalArticle

    Abstract

    The aim of this paper is to present a functional analysis of a diffusion tensor tract statistics (FADTTS) pipeline for delineating the association between multiple diffusion properties along major white matter fiber bundles with a set of covariates of interest, such as age, diagnostic status and gender, and the structure of the variability of these white matter tract properties in various diffusion tensor imaging studies. The FADTTS integrates five statistical tools: (i) a multivariate varying coefficient model for allowing the varying coefficient functions in terms of arc length to characterize the varying associations between fiber bundle diffusion properties and a set of covariates, (ii) a weighted least squares estimation of the varying coefficient functions, (iii) a functional principal component analysis to delineate the structure of the variability in fiber bundle diffusion properties, (iv) a global test statistic to test hypotheses of interest, and (v) a simultaneous confidence band to quantify the uncertainty in the estimated coefficient functions. Simulated data are used to evaluate the finite sample performance of FADTTS. We apply FADTTS to investigate the development of white matter diffusivities along the splenium of the corpus callosum tract and the right internal capsule tract in a clinical study of neurodevelopment. FADTTS can be used to facilitate the understanding of normal brain development, the neural bases of neuropsychiatric disorders, and the joint effects of environmental and genetic factors on white matter fiber bundles. The advantages of FADTTS compared with the other existing approaches are that they are capable of modeling the structured inter-subject variability, testing the joint effects, and constructing their simultaneous confidence bands. However, FADTTS is not crucial for estimation and reduces to the functional analysis method for the single measure.

    Original languageEnglish (US)
    Pages (from-to)1412-1425
    Number of pages14
    JournalNeuroImage
    Volume56
    Issue number3
    DOIs
    StatePublished - Jun 1 2011

    Fingerprint

    Internal Capsule
    Diffusion Tensor Imaging
    Corpus Callosum
    Principal Component Analysis
    Least-Squares Analysis
    Uncertainty
    White Matter
    Brain
    Clinical Studies

    Keywords

    • Confidence band
    • Diffusion tensor imaging
    • Fiber bundle
    • Global test statistic
    • Varying coefficient model

    ASJC Scopus subject areas

    • Cognitive Neuroscience
    • Neurology

    Cite this

    Zhu, H., Kong, L., Li, R., Styner, M., Gerig, G., Lin, W., & Gilmore, J. H. (2011). FADTTS: Functional analysis of diffusion tensor tract statistics. NeuroImage, 56(3), 1412-1425. https://doi.org/10.1016/j.neuroimage.2011.01.075

    FADTTS : Functional analysis of diffusion tensor tract statistics. / Zhu, Hongtu; Kong, Linglong; Li, Runze; Styner, Martin; Gerig, Guido; Lin, Weili; Gilmore, John H.

    In: NeuroImage, Vol. 56, No. 3, 01.06.2011, p. 1412-1425.

    Research output: Contribution to journalArticle

    Zhu, H, Kong, L, Li, R, Styner, M, Gerig, G, Lin, W & Gilmore, JH 2011, 'FADTTS: Functional analysis of diffusion tensor tract statistics', NeuroImage, vol. 56, no. 3, pp. 1412-1425. https://doi.org/10.1016/j.neuroimage.2011.01.075
    Zhu H, Kong L, Li R, Styner M, Gerig G, Lin W et al. FADTTS: Functional analysis of diffusion tensor tract statistics. NeuroImage. 2011 Jun 1;56(3):1412-1425. https://doi.org/10.1016/j.neuroimage.2011.01.075
    Zhu, Hongtu ; Kong, Linglong ; Li, Runze ; Styner, Martin ; Gerig, Guido ; Lin, Weili ; Gilmore, John H. / FADTTS : Functional analysis of diffusion tensor tract statistics. In: NeuroImage. 2011 ; Vol. 56, No. 3. pp. 1412-1425.
    @article{54d128d1e3e842fa8db0b52345a9c357,
    title = "FADTTS: Functional analysis of diffusion tensor tract statistics",
    abstract = "The aim of this paper is to present a functional analysis of a diffusion tensor tract statistics (FADTTS) pipeline for delineating the association between multiple diffusion properties along major white matter fiber bundles with a set of covariates of interest, such as age, diagnostic status and gender, and the structure of the variability of these white matter tract properties in various diffusion tensor imaging studies. The FADTTS integrates five statistical tools: (i) a multivariate varying coefficient model for allowing the varying coefficient functions in terms of arc length to characterize the varying associations between fiber bundle diffusion properties and a set of covariates, (ii) a weighted least squares estimation of the varying coefficient functions, (iii) a functional principal component analysis to delineate the structure of the variability in fiber bundle diffusion properties, (iv) a global test statistic to test hypotheses of interest, and (v) a simultaneous confidence band to quantify the uncertainty in the estimated coefficient functions. Simulated data are used to evaluate the finite sample performance of FADTTS. We apply FADTTS to investigate the development of white matter diffusivities along the splenium of the corpus callosum tract and the right internal capsule tract in a clinical study of neurodevelopment. FADTTS can be used to facilitate the understanding of normal brain development, the neural bases of neuropsychiatric disorders, and the joint effects of environmental and genetic factors on white matter fiber bundles. The advantages of FADTTS compared with the other existing approaches are that they are capable of modeling the structured inter-subject variability, testing the joint effects, and constructing their simultaneous confidence bands. However, FADTTS is not crucial for estimation and reduces to the functional analysis method for the single measure.",
    keywords = "Confidence band, Diffusion tensor imaging, Fiber bundle, Global test statistic, Varying coefficient model",
    author = "Hongtu Zhu and Linglong Kong and Runze Li and Martin Styner and Guido Gerig and Weili Lin and Gilmore, {John H.}",
    year = "2011",
    month = "6",
    day = "1",
    doi = "10.1016/j.neuroimage.2011.01.075",
    language = "English (US)",
    volume = "56",
    pages = "1412--1425",
    journal = "NeuroImage",
    issn = "1053-8119",
    publisher = "Academic Press Inc.",
    number = "3",

    }

    TY - JOUR

    T1 - FADTTS

    T2 - Functional analysis of diffusion tensor tract statistics

    AU - Zhu, Hongtu

    AU - Kong, Linglong

    AU - Li, Runze

    AU - Styner, Martin

    AU - Gerig, Guido

    AU - Lin, Weili

    AU - Gilmore, John H.

    PY - 2011/6/1

    Y1 - 2011/6/1

    N2 - The aim of this paper is to present a functional analysis of a diffusion tensor tract statistics (FADTTS) pipeline for delineating the association between multiple diffusion properties along major white matter fiber bundles with a set of covariates of interest, such as age, diagnostic status and gender, and the structure of the variability of these white matter tract properties in various diffusion tensor imaging studies. The FADTTS integrates five statistical tools: (i) a multivariate varying coefficient model for allowing the varying coefficient functions in terms of arc length to characterize the varying associations between fiber bundle diffusion properties and a set of covariates, (ii) a weighted least squares estimation of the varying coefficient functions, (iii) a functional principal component analysis to delineate the structure of the variability in fiber bundle diffusion properties, (iv) a global test statistic to test hypotheses of interest, and (v) a simultaneous confidence band to quantify the uncertainty in the estimated coefficient functions. Simulated data are used to evaluate the finite sample performance of FADTTS. We apply FADTTS to investigate the development of white matter diffusivities along the splenium of the corpus callosum tract and the right internal capsule tract in a clinical study of neurodevelopment. FADTTS can be used to facilitate the understanding of normal brain development, the neural bases of neuropsychiatric disorders, and the joint effects of environmental and genetic factors on white matter fiber bundles. The advantages of FADTTS compared with the other existing approaches are that they are capable of modeling the structured inter-subject variability, testing the joint effects, and constructing their simultaneous confidence bands. However, FADTTS is not crucial for estimation and reduces to the functional analysis method for the single measure.

    AB - The aim of this paper is to present a functional analysis of a diffusion tensor tract statistics (FADTTS) pipeline for delineating the association between multiple diffusion properties along major white matter fiber bundles with a set of covariates of interest, such as age, diagnostic status and gender, and the structure of the variability of these white matter tract properties in various diffusion tensor imaging studies. The FADTTS integrates five statistical tools: (i) a multivariate varying coefficient model for allowing the varying coefficient functions in terms of arc length to characterize the varying associations between fiber bundle diffusion properties and a set of covariates, (ii) a weighted least squares estimation of the varying coefficient functions, (iii) a functional principal component analysis to delineate the structure of the variability in fiber bundle diffusion properties, (iv) a global test statistic to test hypotheses of interest, and (v) a simultaneous confidence band to quantify the uncertainty in the estimated coefficient functions. Simulated data are used to evaluate the finite sample performance of FADTTS. We apply FADTTS to investigate the development of white matter diffusivities along the splenium of the corpus callosum tract and the right internal capsule tract in a clinical study of neurodevelopment. FADTTS can be used to facilitate the understanding of normal brain development, the neural bases of neuropsychiatric disorders, and the joint effects of environmental and genetic factors on white matter fiber bundles. The advantages of FADTTS compared with the other existing approaches are that they are capable of modeling the structured inter-subject variability, testing the joint effects, and constructing their simultaneous confidence bands. However, FADTTS is not crucial for estimation and reduces to the functional analysis method for the single measure.

    KW - Confidence band

    KW - Diffusion tensor imaging

    KW - Fiber bundle

    KW - Global test statistic

    KW - Varying coefficient model

    UR - http://www.scopus.com/inward/record.url?scp=79955468497&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=79955468497&partnerID=8YFLogxK

    U2 - 10.1016/j.neuroimage.2011.01.075

    DO - 10.1016/j.neuroimage.2011.01.075

    M3 - Article

    C2 - 21335092

    AN - SCOPUS:79955468497

    VL - 56

    SP - 1412

    EP - 1425

    JO - NeuroImage

    JF - NeuroImage

    SN - 1053-8119

    IS - 3

    ER -