Advertisement

Statistical Learning of Spatiotemporal Patterns from Longitudinal Manifold-Valued Networks

  • I. Koval
  • J.-B. Schiratti
  • A. Routier
  • M. Bacci
  • O. Colliot
  • S. Allassonnière
  • S. Durrleman
  • The Alzheimer’s Disease Neuroimaging Initiative
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10433)

Abstract

We introduce a mixed-effects model to learn spatiotemporal patterns on a network by considering longitudinal measures distributed on a fixed graph. The data come from repeated observations of subjects at different time points which take the form of measurement maps distributed on a graph such as an image or a mesh. The model learns a typical group-average trajectory characterizing the propagation of measurement changes across the graph nodes. The subject-specific trajectories are defined via spatial and temporal transformations of the group-average scenario, thus estimating the variability of spatiotemporal patterns within the group. To estimate population and individual model parameters, we adapted a stochastic version of the Expectation-Maximization algorithm, the MCMC-SAEM. The model is used to describe the propagation of cortical atrophy during the course of Alzheimer’s Disease. Model parameters show the variability of this average pattern of atrophy in terms of trajectories across brain regions, age at disease onset and pace of propagation. We show that the personalization of this model yields accurate prediction of maps of cortical thickness in patients.

Notes

Acknowledgements

This work has been partly funded by ERC grant N\(^\mathrm{o}\) 678304, H2020 EU grant N\(^\mathrm{o}\) 666992, and ANR grant ANR-10-IAIHU-06.

References

  1. 1.
    Allassonnière, S., Kuhn, E., Trouvé, A.: Construction of bayesian deformable models via a stochastic approximation algorithm: a convergence study. Bernoulli 16(3), 641–678 (2010)MathSciNetCrossRefMATHGoogle Scholar
  2. 2.
    Atchadé, Y.F.: An adaptive version for the metropolis adjusted langevin algorithm with a truncated drift. Methodol. Comput. Appl. Probab. 8(2), 235–254 (2006)MathSciNetCrossRefMATHGoogle Scholar
  3. 3.
    Bilgel, M., Prince, J.L., Wong, D.F., Resnick, S.M., Jedynak, B.M.: A multivariate nonlinear mixed effects model for longitudinal image analysis: application to amyloid imaging. NeuroImage 134, 658–670 (2016)CrossRefGoogle Scholar
  4. 4.
    Delyon, B., Lavielle, M., Moulines, E.: Convergence of a stochastic approximation version of the EM algorithm. Ann. Stat. 27, 94–128 (1999)MathSciNetCrossRefMATHGoogle Scholar
  5. 5.
    Do Carmo Valero, M.P.: Riemannian Geometry. Birkhäuser, Basel (1992)CrossRefGoogle Scholar
  6. 6.
    Donohue, M., Jacqmin-Gadda, H., Goff, M.L., Thomas, R., Raman, R., Gams, A., Beckett, L., Jack, C., Weiner, M., Dartigues, J.F., Aisen, P., The ADNI: Estimating long-term multivariate progression from short-term data. Alzheimer’s Dement. 10(5), 400–410 (2014)CrossRefGoogle Scholar
  7. 7.
    Guerrero, R., Schmidt-Richberg, A., Ledig, C., Tong, T., Wolz, R., Rueckert, D., The ADNI: Instantiated mixed effects modeling of alzheimer’s disease markers. Neuroimage 142, 113–125 (2016)CrossRefGoogle Scholar
  8. 8.
    Hampel, H., O’Bryant, S.E., Durrleman, S., Younesi, E., Rojkova, K., Escott-Price, V., Corvol, J.C., Broich, K., Dubois, B., Lista, S., The Alzheimer Precision Medicine Initiative: A precision medicine initiative for alzheimer’s disease: the road ahead to biomarker-guided integrative disease modeling. Climacteric 20(2), 107–118 (2017)CrossRefGoogle Scholar
  9. 9.
    Iturria-Medina, Y., Sotero, R., Toussaint, P., Mateos-Pérez, J., Evans, A., The ADNI: Early role of vascular dysregulation on late-onset alzheimer’s disease based on multifactorial data-driven analysis. Nat. Commun. 7, 11934 (2016)CrossRefGoogle Scholar
  10. 10.
    Kuhn, E., Lavielle, M.: Coupling a stochastic approximation version of EM with an MCMC procedure. ESAIM Probab. Stat. 8, 115–131 (2004)MathSciNetCrossRefMATHGoogle Scholar
  11. 11.
    Kuhn, E., Lavielle, M.: Maximum likelihood estimation in nonlinear mixed effects models. Comput. Stat. Data Anal. 49(4), 1020–1038 (2005)MathSciNetCrossRefMATHGoogle Scholar
  12. 12.
    Lee, J.M.: Smooth manifolds. In: Lee, J.M. (ed.) Introduction to Smooth Manifolds. Graduate Texts in Mathematics, vol. 218, pp. 1–29. Springer, New York (2003). doi: 10.1007/978-0-387-21752-9_1 CrossRefGoogle Scholar
  13. 13.
    Raj, A., Kuceyeski, A., Weiner, M.: A network diffusion model of disease progression in dementia. Neuron 73, 1204–1215 (2012)CrossRefGoogle Scholar
  14. 14.
    Reuter, M., Schmandsky, N., Rosas, H., Fischl, B.: Within-subject template estimation for unbiased longitudinal image analysis. Neuroimage 61(4), 1402–1418 (2012)CrossRefGoogle Scholar
  15. 15.
    Schiratti, J.B., Allassonnière, S., Colliot, O., Durrleman, S.: Learning spatiotemporal trajectories from manifold-valued longitudinal data. In: Advances in Neural Information Processing Systems, pp. 2404–2412 (2015)Google Scholar
  16. 16.
    Young, A.L., et al.: Multiple orderings of events in disease progression. In: Ourselin, S., Alexander, D.C., Westin, C.-F., Cardoso, M.J. (eds.) IPMI 2015. LNCS, vol. 9123, pp. 711–722. Springer, Cham (2015). doi: 10.1007/978-3-319-19992-4_56 CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • I. Koval
    • 1
    • 3
  • J.-B. Schiratti
    • 1
    • 3
  • A. Routier
    • 1
  • M. Bacci
    • 1
  • O. Colliot
    • 1
    • 2
  • S. Allassonnière
    • 3
  • S. Durrleman
    • 1
  • The Alzheimer’s Disease Neuroimaging Initiative
  1. 1.Inria Paris-Rocquencourt, Inserm U1127, CNRS UMR 7225, Sorbonne Universités, UPMC Univ Paris 06 UMRS 1127, Institut du Cerveau et de la Moelle épinière, ICMParisFrance
  2. 2.AP-HP, Pitié-Salpêtrière, Neuroradiology DepartmentParisFrance
  3. 3.INSERM UMRS 1138, Centre de Recherche des Cordeliers, Université Paris DescartesParisFrance

Personalised recommendations