Label-Informed Non-negative Matrix Factorization with Manifold Regularization for Discriminative Subnetwork Detection

  • Takanori Watanabe
  • Birkan Tunc
  • Drew Parker
  • Junghoon Kim
  • Ragini Verma
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9900)


In this paper, we present a novel method for obtaining a low dimensional representation of a complex brain network that: (1) can be interpreted in a neurobiologically meaningful way, (2) emphasizes group differences by accounting for label information, and (3) captures the variation in disease subtypes/severity by respecting the intrinsic manifold structure underlying the data. Our method is a supervised variant of non-negative matrix factorization (NMF), and achieves dimensionality reduction by extracting an orthogonal set of subnetworks that are interpretable, reconstructive of the original data, and also discriminative at the group level. In addition, the method includes a manifold regularizer that encourages the low dimensional representations to be smooth with respect to the intrinsic geometry of the data, allowing subjects with similar disease-severity to share similar network representations. While the method is generalizable to other types of non-negative network data, in this work we have used structural connectomes (SCs) derived from diffusion data to identify the cortical/subcortical connections that have been disrupted in abnormal neurological state. Experiments on a traumatic brain injury (TBI) dataset demonstrate that our method can identify subnetworks that can reliably classify TBI from controls and also reveal insightful connectivity patterns that may be indicative of a biomarker.


Support Vector Machine Traumatic Brain Injury Independent Component Analysis Alternate Direction Method Manifold Regularization 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Abraham, A., Pedregosa, F., Eickenberg, M., Gervais, P., Mueller, A., et al.: Machine learning for neuroimaging with scikit-learn. Front. Neuroinformatics 8(14) (2014)Google Scholar
  2. 2.
    Allahyar, A., Ridder, J.: FERAL: network-based classifier with application to breast cancer outcome prediction. Bioinformatics 31(12), i311–i319 (2015)CrossRefGoogle Scholar
  3. 3.
    Behrens, T., et al.: Non-invasive mapping of connections between human thalamus and cortex using diffusion imaging. Nat. Neurosci. 6(7), 750–757 (2003)CrossRefGoogle Scholar
  4. 4.
    Boutsidis, C., Gallopoulos, E.: SVD based initialization: a head start for nonnegative matrix factorization. Pattern Recognit. 41, 1350–1362 (2008)CrossRefzbMATHGoogle Scholar
  5. 5.
    Cheplygina, V., Tax, D.M., Loog, M., Feragen, A.: Network-guided group feature selection for classification of autism spectrum disorder. In: Wu, G., Zhang, D., Zhou, L. (eds.) MLMI 2014. LNCS, vol. 8679, pp. 190–197. Springer, Heidelberg (2014)Google Scholar
  6. 6.
    Fan, R.E., Chang, K.W., Hsieh, C.J., Wang, X.R., Lin, C.J.: LIBLINEAR: a library for large linear classification. J. Mach. Learn. Res. 9, 1871–1874 (2008)zbMATHGoogle Scholar
  7. 7.
    Ghanbari, Y., Smith, A.R., Schultz, R.T., Verma, R.: Identifying group discriminative and age regressive sub-networks from DTI-based connectivity via a unified framework of non-negative matrix factorization and graph embedding. Med. Image Anal. 18(8) (2014)Google Scholar
  8. 8.
    Kasenburg, N., et al.: Supervised hub-detection for brain connectivity. In: Proceedings of the SPIE, vol. 9784, Medical Imaging 2016: Image Processing, p. 978409 (2016)Google Scholar
  9. 9.
    Lee, D.D., Seung, H.S.: Learning the parts of objects by NMF. Nature 401, 788–791 (1999)CrossRefGoogle Scholar
  10. 10.
    Liu, X., et al., H.: Projective nonnegative graph embedding. IEEE Trans. Image Proc. (2010)Google Scholar
  11. 11.
    Manton, J.H.: Optimization algorithms exploiting unitary constraints. IEEE Trans. Signal Process. 50(3), 635–650 (2002)MathSciNetCrossRefGoogle Scholar
  12. 12.
    Xu, Y., Yin, W., Wen, Z., Zhang, Y.: An alternating direction algorithm for matrix completion with nonnegative factors. Front. Math. China 7(2), 365–384 (2012)MathSciNetCrossRefzbMATHGoogle Scholar

Copyright information

© Springer International Publishing AG 2016

Authors and Affiliations

  • Takanori Watanabe
    • 1
  • Birkan Tunc
    • 1
  • Drew Parker
    • 1
  • Junghoon Kim
    • 2
  • Ragini Verma
    • 1
  1. 1.Section of Biomedical Image AnalysisUniversity of PennsylvaniaPhiladelphiaUSA
  2. 2.The City College of New YorkNew YorkUSA

Personalised recommendations