Predicting Cognitive Data from Medical Images Using Sparse Linear Regression

  • Benjamin M. Kandel
  • David A. Wolk
  • James C. Gee
  • Brian Avants
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7917)

Abstract

We present a new framework for predicting cognitive or other continuous-variable data from medical images. Current methods of probing the connection between medical images and other clinical data typically use voxel-based mass univariate approaches. These approaches do not take into account the multivariate, network-based interactions between the various areas of the brain and do not give readily interpretable metrics that describe how strongly cognitive function is related to neuroanatomical structure. On the other hand, high-dimensional machine learning techniques do not typically provide a direct method for discovering which parts of the brain are used for making predictions. We present a framework, based on recent work in sparse linear regression, that addresses both drawbacks of mass univariate approaches, while preserving the direct spatial interpretability that they provide. In addition, we present a novel optimization algorithm that adapts the conjugate gradient method for sparse regression on medical imaging data. This algorithm produces coefficients that are more interpretable than existing sparse regression techniques.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Ashburner, J., Friston, K.J.: Voxel-based morphometry–the methods. NeuroImage 11(6 pt. 1), 805–821 (2000); PMID: 10860804Google Scholar
  2. 2.
    Ashburner, J.: A fast diffeomorphic image registration algorithm. NeuroImage 38(1), 95–113 (2007)CrossRefGoogle Scholar
  3. 3.
    Avants, B.B., Epstein, C.L., Grossman, M., Gee, J.C.: Symmetric diffeomorphic image registration with cross-correlation: Evaluating automated labeling of elderly and neurodegenerative brain. Medical Image Analysis 12(1), 26–41 (2008)CrossRefGoogle Scholar
  4. 4.
    Batmanghelich, N., Taskar, B., Davatzikos, C.: A general and unifying framework for feature construction, in image-based pattern classification. In: Prince, J.L., Pham, D.L., Myers, K.J. (eds.) IPMI 2009. LNCS, vol. 5636, pp. 423–434. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  5. 5.
    Bertsekas, D.: On the goldstein-levitin-polyak gradient projection method. IEEE Transactions on Automatic Control 21(2), 174–184 (1976)MathSciNetMATHCrossRefGoogle Scholar
  6. 6.
    Candès, E.J., Romberg, J., Tao, T.: Robust uncertainty principles: Exact signal reconstruction from highly incomplete frequency information. IEEE Transactions on Information Theory 52(2), 489–509 (2006)MATHCrossRefGoogle Scholar
  7. 7.
    Das, S.R., Avants, B.B., Grossman, M., Gee, J.C.: Registration based cortical thickness measurement. NeuroImage 45(3), 867–879 (2009); PMID: 19150502Google Scholar
  8. 8.
    Davatzikos, C., Resnick, S.M., Wu, X., Parmpi, P., Clark, C.M.: Individual patient diagnosis of AD and FTD via high-dimensional pattern classification of MRI. NeuroImage 41(4), 1220–1227 (2008)CrossRefGoogle Scholar
  9. 9.
    Davatzikos, C.: Why voxel-based morphometric analysis should be used with great caution when characterizing group differences. NeuroImage 23(1), 17–20 (2004)CrossRefGoogle Scholar
  10. 10.
    Donoho, D.L.: De-noising by soft-thresholding. IEEE Transactions on Information Theory 41(3), 613–627 (1995)MathSciNetMATHCrossRefGoogle Scholar
  11. 11.
    Donoho, D.L.: For most large underdetermined systems of linear equations the minimal l1-norm solution is also the sparsest solution. Communications on Pure and Applied Mathematics 59(6), 797–829 (2006)MathSciNetMATHCrossRefGoogle Scholar
  12. 12.
    Efron, B., Hastie, T., Johnstone, I., Tibshirani, R.: Least angle regression. The Annals of Statistics 32(2), 407–499 (2004)MathSciNetMATHCrossRefGoogle Scholar
  13. 13.
    Franke, K., Ziegler, G., Klöppel, S., Gaser, C.: Estimating the age of healthy subjects from t1-weighted MRI scans using kernel methods: Exploring the influence of various parameters. NeuroImage 50(3), 883–892 (2010)CrossRefGoogle Scholar
  14. 14.
    Friedman, J., Hastie, T., Tibshirani, R.: Regularization paths for generalized linear models via coordinate descent. Journal of Statistical Software 33(1), 1 (2010)Google Scholar
  15. 15.
    Ganesh, G., Burdet, E., Haruno, M., Kawato, M.: Sparse linear regression for reconstructing muscle activity from human cortical fMRI. NeuroImage 42(4), 1463–1472 (2008)CrossRefGoogle Scholar
  16. 16.
    Goldstein, A.A.: Convex programming in Hilbert space. Bulletin of the American Mathematical Society 70(5), 709–710 (1964)MathSciNetMATHCrossRefGoogle Scholar
  17. 17.
    Lee, H., Lee, D.S., Kang, H., Kim, B.-N., Chung, M.K.: Sparse brain network recovery under compressed sensing. IEEE Transactions on Medical Imaging 30(5), 1154–1165Google Scholar
  18. 18.
    Morris, J.C., Heyman, A., Mohs, R.C., Hughes, J.P., van Belle, G., Fillenbaum, G., Mellits, E.D., Clark, C.: The consortium to establish a registry for Alzheimer’s disease (CERAD). Part i. Clinical and neuropsychological assessment of Alzheimer’s disease. Neurology 39(9), 1159–1165 (1989); PMID: 2771064Google Scholar
  19. 19.
    Natarajan, B.K.: Sparse approximate solutions to linear systems. SIAM Journal on Computing 24(2), 227–234 (1995)MathSciNetMATHCrossRefGoogle Scholar
  20. 20.
    Schwartz, A., Polak, E.: Family of projected descent methods for optimization problems with simple bounds. Journal of Optimization Theory and Applications 92(1), 1–31 (1997)MathSciNetMATHCrossRefGoogle Scholar
  21. 21.
    Schwarz, G.: Estimating the dimension of a model. The Annals of Statistics 6(2), 461–464 (1978)MathSciNetMATHCrossRefGoogle Scholar
  22. 22.
    Tibshirani, R.: Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society. Series B (Methodological) 58(1), 267–288 (1996)MathSciNetMATHGoogle Scholar
  23. 23.
    Tibshirani, R., Saunders, M., Rosset, S., Zhu, J., Knight, K.: Sparsity and smoothness via the fused lasso. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 67(1), 91–108 (2005)MathSciNetMATHCrossRefGoogle Scholar
  24. 24.
    Tropp, J.A., Gilbert, A.C.: Signal recovery from random measurements via orthogonal matching pursuit. IEEE Transactions on Information Theory 53(12), 4655–4666 (2007)MathSciNetCrossRefGoogle Scholar
  25. 25.
    Tropp, J.A., Wright, S.J.: Computational methods for sparse solution of linear inverse problems. Proceedings of the IEEE 98(6), 948–958 (2010)CrossRefGoogle Scholar
  26. 26.
    Wang, B., Pham, T.D.: MRI-based age prediction using hidden markov models. Journal of Neuroscience Methods 199(1), 140–145 (2011)CrossRefGoogle Scholar
  27. 27.
    Witten, D.M., Tibshirani, R., Hastie, T.: A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis. Biostatistics 10(3), 515–534 (1937); PMID: 19377034Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Benjamin M. Kandel
    • 1
    • 4
  • David A. Wolk
    • 2
  • James C. Gee
    • 3
    • 4
  • Brian Avants
    • 3
    • 4
  1. 1.Department of BioengineeringUniversity of PennsylvaniaUSA
  2. 2.Department of Neurology and Penn Memory CenterUniversity of PennsylvaniaUSA
  3. 3.Department of RadiologyUniversity of PennsylvaniaUSA
  4. 4.Penn Image Computing and Science Laboratory (PICSL)USA

Personalised recommendations