Compression of fMRI data using wavelet tensor train decomposition

Authors

  • P.V. Kharyuk
  • I.V. Оseledets
  • V.L. Ushakov

Keywords:

numerical tensor methods
Daubechies wavelet transform
wavelet tensor train decomposition
functional magnetic resonance imaging (fMRI) data
lossy data compression

Abstract

The application of the Wavelet Tensor Train (WTT) decomposition to the compression of functional magnetic resonance imaging (fMRI) data is considered. Contrary to the classical wavelet transforms, the WTT decomposition is an algebraic technique for the construction of adaptive wavelet transforms, but it requires to store filters for each data array. The WTT method of compressing realistic fMRI data is compared with Daubechies wavelet transforms. The numerical results show that the WTT transform can be successfully used to compress lossy data.

New computational technologies

Downloads

Published

2014-12-07

Issue

Vol. 15 (2014): Issue 4.

Section

Section 1. Numerical methods and applications

Author Biographies

P.V. Kharyuk

Lomonosov Moscow State University,
Faculty of Computational Mathematics and Cybernetics
• PhD Student

I.V. Оseledets

Institute of Numerical Mathematics of RAS (INM RAS)
• Senior Researcher

V.L. Ushakov

National Research Center «Kurchatov Institute»
• Head of Laboratory

References

  1. Liu Y., Pearlman W.A. Four-dimensional wavelet compression of 4-D medical images using scalable 4-D SBHP // Proc. Data Compression Conference. Washington DC: IEEE Press, 2007. 233-242.
  2. Lalgudi H.G., Bilgin A., Marcellin M.W., et al. Four-dimensional compression of fMRI using JPEG2000 // SPIE Proc. 2005. 5747. 1028-1037.
  3. Rajeswari R., Rajesh R. Efficient compression of 4D fMRI images using bandelet transform and fuzzy thresholding // Proc. World Congress on Nature and Biologically Inspired Computing. New York: IEEE Press, 2009. 543-547.
  4. Knezovic J., Kovavc M., vZagar M., et al. Novel prediction based technique for efficient compression of medical imaging data // Telemedicine Techniques and Applications. Rijeka: InTech Press, 2011. 169-184.
  5. Ruchika M., Singh M., Singh A.R. Compression of medical images using wavelet transforms // Int. J. of Soft Computing and Engineering. 2012. 2, N 2. 339-343.
  6. Brechet L., Lucas M.-F., Doncarli C., Farina D. Compression of biomedical signals with mother wavelet optimization and best-basis wavelet packet selection // IEEE Trans. on Biomedical Engineering. 2007. 54, N 12. 2186-2192.
  7. Nielsen M., Kamavuako E.N., Andersen M.M., et al. Optimal wavelets for biomedical signal compression // Medical and Biological Engineering and Computing. 2006. 44, N 7. 561-568.
  8. Nguyen B.P., Chui C.-K., Ong S.-H., Chang S. An efficient compression scheme for 4-D medical images using hierarchical vector quantization and motion compensation // Computers in Biology and Medicine. 2011. 41, N 9. 843-856.
  9. Anusuya V., Raghavan V.S., Kavitha G. Lossless compression on MRI images using SWT // Journal of Digital Imaging. 2014. 27, N 5. 594-600.
  10. Taquet J., Labit C. Hierarchical oriented predictions for resolution scalable lossless and near-lossless compression of CT and MRI biomedical images // IEEE Trans. on Image Processing. 2012. 21, N 5. 2641-2652.
  11. Pizzolante R., Carpentieri B. Lossless, low-complexity, compression of three-dimensional volumetric medical images via linear prediction // Proc. 18th Int. Conf. on Digital Signal Processing. New York: IEEE Press, 2013. 1-6.
  12. Ogawa S., Lee T.M., Kay A.R., Tank D.W. Brain magnetic resonance imaging with contrast dependent on blood oxygenation // Proc. of the National Academy of Sciences of USA. 1990. 87, N 24. 9868-9872.
  13. Mulert C., Lemieux L. EEG - fMRI: physiological basis, technique, and applications. Heidelberg: Springer, 2010.
  14. Daubechies I. Ten lectures on wavelets. Philadelphia: SIAM, 1992.
  15. Oseledets I.V., Tyrtyshnikov E.E. Algebraic wavelet transform via quantics tensor train decomposition // SIAM J. Sci. Comput. 2011. 33, N 3. 1315-1328.
  16. Kazeev V.A., Oseledets I.V. The tensor structure of a class of adaptive algebraic wavelet transforms. Preprint N 2013-28. Zürich: Swiss Federal Inst. of Technology, 2013.
  17. Oseledets I.V. Tensor-train decomposition // SIAM J. Sci. Comput. 2011. 33, N 5. 2295-2317.
  18. Oseledets I.V. Approximation of 2^d imes 2^d matrices using tensor decomposition // SIAM J. Matrix Anal. Appl. 2010. 31, N 4. 2130-2145.
  19. Khoromskij B.N. O(d log N)-quantics approximation of N-d tensors in high-dimensional numerical modeling // Constr. Appr. 2011. 34, N 2. 257-280.
  20. Харюк П.В., Оселедец И.В. WTT-разложение для семейств массивов и его применение для сжатия изображений // Вычислительные методы и программирование. 2014. 15. 229-238.
  21. Theis F.J., Tanaka T. A fast and efficient method for compressing fMRI data sets // Lecture Notes in Computer Science. Vol. 3697. Heidelberg: Springer, 2005. 769-777.
  22. Taswell C. Wavelet transform compression of functional magnetic resonance image sequences // Proc. Int. Conf. on Signal and Image Processing. 1998. http://www.toolsmiths.com/docs/sip98.pdf.