Joint Sparse and Low-Rank Regularized Multi-Task Multi-Linear Regression for Prediction of Infant Brain Development with Incomplete Data

  • Ehsan Adeli
  • Yu Meng
  • Gang Li
  • Weili Lin
  • Dinggang ShenEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10433)


Studies involving dynamic infant brain development has received increasing attention in the past few years. For such studies, a complete longitudinal dataset is often required to precisely chart the early brain developmental trajectories. Whereas, in practice, we often face missing data at different time point(s) for different subjects. In this paper, we propose a new method for prediction of infant brain development scores at future time points based on longitudinal imaging measures at early time points with possible missing data. We treat this as a multi-dimensional regression problem, for predicting multiple brain development scores (multi-task) from multiple previous time points (multi-linear). To solve this problem, we propose an objective function with a joint \(\ell _1\) and low-rank regularization on the mapping weight tensor, to enforce feature selection, while preserving the structural information from multiple dimensions. Also, based on the bag-of-words model, we propose to extract features from longitudinal imaging data. The experimental results reveal that we can effectively predict the brain development scores assessed at the age of four years, using the imaging data as early as two years of age.


  1. 1.
    Boyd, S., et al.: Distributed optimization and statistical learning via the alternating direction method of multipliers. Found. Trends Mach. Learn. 3(1), 1–122 (2011)Google Scholar
  2. 2.
    Cai, J.F., Candès, E., Shen, Z.: A singular value thresholding algorithm for matrix completion. SIAM J. Optim. 20(4), 1956–1982 (2010)MathSciNetCrossRefzbMATHGoogle Scholar
  3. 3.
    Caruana, R.: Multitask learning. In: Thrun, S., Pratt, L. (eds.) Learning to Learn, pp. 95–133. Springer, New York (1998)Google Scholar
  4. 4.
    Eckstein, J., Yao, W.: Understanding the convergence of the alternating direction method of multipliers: theoretical and computational perspectives. Pac. J. Optim. 11(4), 619–644 (2015)Google Scholar
  5. 5.
    Gaiffas, S., Lecué, G.: Sharp oracle inequalities for high-dimensional matrix prediction. IEEE Trans. Inf. Theor. 57(10), 6942–6957 (2011)MathSciNetCrossRefzbMATHGoogle Scholar
  6. 6.
    Li, G., et al.: Mapping longitudinal development of local cortical gyrification in infants from birth to 2 years of age. J. Neurosci. 34(12), 4228–4238 (2014)Google Scholar
  7. 7.
    Li, G., et al.: Construction of 4D high-definition cortical surface atlases of infants: methods and applications. Med. Image Anal. 25(1), 22–36 (2015)Google Scholar
  8. 8.
    Meng, Y., et al.: Learning-based subject-specific estimation of dynamic maps of cortical morphology at missing time points in longitudinal infant studies. Hum. Brain Mapp. 37(11), 4129–4147 (2016)Google Scholar
  9. 9.
    Mosabbeb, E.A., et al.: Robust feature-sample linear discriminant analysis for brain disorders diagnosis. In: NIPS, pp. 658–666 (2015)Google Scholar
  10. 10.
    Romera-Paredes, B., Aung, H., Bianchi-Berthouze, N., Pontil, M.: Multilinear multitask learning. In: ICML, pp. 1444–1452 (2013)Google Scholar
  11. 11.
    Sivic, J., Zisserman, A.: Efficient visual search of videos cast as text retrieval. IEEE TPAMI 31(4), 591–606 (2009)CrossRefGoogle Scholar
  12. 12.
    Song, X., Lu, H.: Multilinear regression for embedded feature selection with application to FMRI analysis. In: AAAI (2016)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Ehsan Adeli
    • 1
  • Yu Meng
    • 1
  • Gang Li
    • 1
  • Weili Lin
    • 1
  • Dinggang Shen
    • 1
    Email author
  1. 1.Department of Radiology and BRICUniversity of North Carolina at Chapel HillChapel HillUSA

Personalised recommendations