Trajectories from Distribution-Valued Functional Curves: A Unified Wasserstein Framework

Temporal changes in medical images are often evaluated along a parametrized function that represents a structure of interest (e.g. white matter tracts). By attributing samples along these functions with distributions of image properties in the local neighborhood, we create distribution-valued signatures for these functions. We propose a novel, comprehensive framework which models their temporal evolution trajectories. This is achieved under the unifying scheme of Wasserstein distance metric. The regression problem is formulated as a constrained optimization problem and solved using an alternating projection algorithm. The solution simultaneously preserves the functional characteristics of the curve, models the temporal change in distribution profiles and forces the estimated distributions to be valid. Hypothesis testing is applied in two ways using Wasserstein based test statistics. Validation is presented on synthetic data. Estimation of a population trajectory is shown using diffusion properties along DTI tracts from a healthy population of infants. Detection of delayed growth is shown using a case study.

[1]  Martin Styner,et al.  Spatiotemporal modeling of distribution-valued data applied to DTI tract evolution in infant neurodevelopment , 2013, 2013 IEEE 10th International Symposium on Biomedical Imaging.

[2]  David B. Dunson,et al.  Mapping population-based structural connectomes , 2018, NeuroImage.

[3]  Anuj Srivastava,et al.  A Novel Representation for Riemannian Analysis of Elastic Curves in Rn , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[4]  Tommi S. Jaakkola,et al.  Modeling Persistent Trends in Distributions , 2015, Journal of the American Statistical Association.

[5]  Aasa Feragen,et al.  Learning from uncertain curves: The 2-Wasserstein metric for Gaussian processes , 2017, NIPS.

[6]  M. C. Jones,et al.  Local Linear Quantile Regression , 1998 .

[7]  Antonio Irpino,et al.  Linear regression for numeric symbolic variables: a least squares approach based on Wasserstein Distance , 2012, Advances in Data Analysis and Classification.

[8]  Ciprian M Crainiceanu,et al.  Longitudinal penalized functional regression for cognitive outcomes on neuronal tract measurements , 2012, Journal of the Royal Statistical Society. Series C, Applied statistics.

[9]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[10]  Anuj Srivastava,et al.  Functional and Shape Data Analysis , 2016 .

[11]  Gabriel Peyré,et al.  Wasserstein barycentric coordinates , 2016, ACM Trans. Graph..

[12]  H. Bondell,et al.  Noncrossing quantile regression curve estimation. , 2010, Biometrika.

[13]  R. Dykstra An Algorithm for Restricted Least Squares Regression , 1983 .

[14]  Arthur F. Kramer,et al.  Discovery and visualization of structural biomarkers from MRI using transport-based morphometry , 2017, NeuroImage.

[15]  A. Irpino,et al.  F eb 2 01 2 Linear regression for numeric symbolic variables : an ordinary least squares approach based on Wasserstein Distance , 2022 .

[16]  Martin Styner,et al.  UNC-Utah NA-MIC framework for DTI fiber tract analysis , 2014, Front. Neuroinform..

[17]  Martin Styner,et al.  Parametric regression scheme for distributions: Analysis of DTI fiber tract diffusion changes in early brain development , 2014, 2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI).

[18]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[19]  Martin Styner,et al.  FADTTS: Functional analysis of diffusion tensor tract statistics , 2011, NeuroImage.

[20]  Jianqing Fan,et al.  Estimation of conditional densities and sensitivity measures in nonlinear dynamical systems , 1996 .

[21]  John H. Gilmore,et al.  Longitudinal growth modeling of discrete-time functions with application to DTI tract evolution in early neurodevelopment , 2012, 2012 9th IEEE International Symposium on Biomedical Imaging (ISBI).

[22]  Gabriel Peyré,et al.  Computational Optimal Transport , 2018, Found. Trends Mach. Learn..

[23]  Jan de Leeuw,et al.  Correctness of Kruskal's algorithms for monotone regression with ties , 1977 .

[24]  J. O. Ramsay,et al.  Functional Data Analysis (Springer Series in Statistics) , 1997 .

[25]  Anuj Srivastava,et al.  Shape Analysis of Elastic Curves in Euclidean Spaces , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[26]  R. Dykstra,et al.  A Method for Finding Projections onto the Intersection of Convex Sets in Hilbert Spaces , 1986 .