TISSUE TYPE DIFFERENTIATION FOR BRAIN GLIOMA USING NON-NEGATIVE MATRIX FACTORIZATION

Yuqian Li, Diana M. Sima, Sofie Van Cauter, Uwe Himmelreich, Yiming Pi, Sabine Van Huffel

2012

Abstract

The purpose of this paper is to introduce a hierarchical Non-negative Matrix Factorization (NMF) approach, customized for the problem of blindly separating brain glioma tumor tissue types using short-echo time proton magnetic resonance spectroscopic imaging (1H MRSI) signals. The proposed algorithm consists of two levels of NMF, where two constituent spectra are computed in each level. The first level is able to correctly detect the spectral profile corresponding to the most predominant tissue type, i.e., normal tissue, while the second level is optimized in order to detect two ‘abnormal’ spectral profiles so that the 3 recovered spectral profiles are least correlated with each other. The two-level decomposition is followed by the reestimation of the overall spatial distribution of each tissue type via standard Non-negative Least Square (NNLS). This method is demonstrated on in vivo short-TE 1H MRSI brain data of a glioblastoma multiforme patient and a grade II-III glioma patient. The results show the possibility of differentiating normal tissue, tumor tissue and necrotic tissue in the form of recovered tissue-specific spectra with accurate spatial distributions.

References

  1. Bottomley PA, 1987. Spatial localization in NMRspectroscopy in vivo. Ann N Y Acad Sci, 508:333-348.
  2. Brown T, Kincaid B, Ugurbil K, 1982. NMR chemical shift imaging in three dimensions. PNAS, 79(11): 3523-3526.
  3. Croitor Sava A, Martinez-Bisbal MC, Van Huffel S, Cerda JM, Sima DM, Celda B, 2011. Ex vivo high resolution magic angle spinning metabolic profiles describe intratumoral histopathological tissue properties in adult human gliomas. Magn Reson Med, 65:320-328.
  4. Cichocki A, Zdunek R, Amari S, 2007. Hierarchical ALS algorithms for nonnegative matrix and 3D tensor factorization. Lect Notes Comput Sci, 4666:169-176.
  5. Cichocki A, Phan AH, 2009. Fast local algorithms for large scale Nonegative Matrix and Tensor Factorizations. IEICE Trans Fund Electron Comm Comput Sci, 3:708-721.
  6. Du S, Mao X, Sajda P and Shungu DC, 2008. Automated tissue segmentation and blind recovery of 1H MRS imaging spectral patterns of normal and diseased human brain. NMR in Biomedicine, 21:33-41.
  7. Gillis N, 2011. Nonnegative matrix factorization complexity, algorithms and applications. PhD Thesis, Louvan-La-Neuve, Belgium.
  8. Howe FA, Barton SJ, Cudlip SA, Stubbs M, Saunders DE, Murphy JR, Opstad KS, Doyle VL, McLean MA, Bell BA, Griffiths JR, 2003. Metabolic profiles of human brain tumours using quantitative in vivo 1H magnetic resonance spectroscopy. Magn Res Med, 49:223-232.
  9. Laudadio T, Mastronardi N, Vanhamme L, Van Hecke P, Van Huffel S, 2002. Improved Lanczos algorithms for blackbox MRS data quantitation. J Magn Reson, 157: 292-297.
  10. Lawson C.L. and R.J. Hanson, 1974, Solving LeastSquares Problems, Prentice-Hall, Chapter 23, p. 161,.
  11. Lee DD, Seung HS, 1999. Learning the parts of objects by non-negative matrix factorization. Nature, 401:788- 791.
  12. Paatero P, Tapper U, 1994. Positive matrix factorization: A non-negative factor model with optimal utilization of error. Environmetrics, 5:111-126.
  13. Poullet JB, 2008. Quantification and classification of magnetic resonance spectroscopic data for brain tumor diagnosis. PhD Thesis, Leuven, Belgium.
  14. Sajda P, Du S, Brown TR, Parra LC, Stoyanova R, 2003. Recovery of constituent spectra in 3D chemical shift imagin using non-negative matrix factorization. In Proc. 4th Int. Symp. ICA and BSS, 71-76.
  15. Sajda P, Du S, Brown TR, Stoyanova R, Shungu DC, Mao X, Parra LC, 2004. Nonnegative Matrix Factorization for Rapid Recovery of Constituent Spectra in Magnetic Resonance Chemical Shift Imaging of the Brain. IEEE Trans Med Imaging, 23:1453-1465.
  16. Su Y, Thakur SB., Karimi S, Du S, Sajda P, Huang W, Parra LC, 2008. Spectrum separation resolves partialvolume effect of MRSI as demonstrated on brain tumor scans. NMR Biomed, 21:1030-1042.
Download


Paper Citation


in Harvard Style

Li Y., Sima D., Van Cauter S., Himmelreich U., Pi Y. and Van Huffel S. (2012). TISSUE TYPE DIFFERENTIATION FOR BRAIN GLIOMA USING NON-NEGATIVE MATRIX FACTORIZATION . In Proceedings of the International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2012) ISBN 978-989-8425-89-8, pages 25-31. DOI: 10.5220/0003734600250031


in Bibtex Style

@conference{biosignals12,
author={Yuqian Li and Diana M. Sima and Sofie Van Cauter and Uwe Himmelreich and Yiming Pi and Sabine Van Huffel},
title={TISSUE TYPE DIFFERENTIATION FOR BRAIN GLIOMA USING NON-NEGATIVE MATRIX FACTORIZATION},
booktitle={Proceedings of the International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2012)},
year={2012},
pages={25-31},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003734600250031},
isbn={978-989-8425-89-8},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2012)
TI - TISSUE TYPE DIFFERENTIATION FOR BRAIN GLIOMA USING NON-NEGATIVE MATRIX FACTORIZATION
SN - 978-989-8425-89-8
AU - Li Y.
AU - Sima D.
AU - Van Cauter S.
AU - Himmelreich U.
AU - Pi Y.
AU - Van Huffel S.
PY - 2012
SP - 25
EP - 31
DO - 10.5220/0003734600250031