The successive projection algorithm as an initialization method for brain tumor segmentation using non-negative matrix factorization

Non-negative matrix factorization (NMF) has become a widely used tool for additive parts-based analysis in a wide range of applications. As NMF is a non-convex problem, the quality of the solution will depend on the initialization of the factor matrices. In this study, the successive projection algorithm (SPA) is proposed as an initialization method for NMF. SPA builds on convex geometry and allocates endmembers based on successive orthogonal subspace projections of the input data. SPA is a fast and reproducible method, and it aligns well with the assumptions made in near-separable NMF analyses. SPA was applied to multi-parametric magnetic resonance imaging (MRI) datasets for brain tumor segmentation using different NMF algorithms. Comparison with common initialization methods shows that SPA achieves similar segmentation quality and it is competitive in terms of convergence rate. Whereas SPA was previously applied as a direct endmember extraction tool, we have shown improved segmentation results when using SPA as an initialization method, as it allows further enhancement of the sources during the NMF iterative procedure.

[1]  M. C. U. Araújo,et al.  The successive projections algorithm for variable selection in spectroscopic multicomponent analysis , 2001 .

[2]  D K Smith,et al.  Numerical Optimization , 2001, J. Oper. Res. Soc..

[3]  Nicolas Gillis,et al.  Accelerated Multiplicative Updates and Hierarchical ALS Algorithms for Nonnegative Matrix Factorization , 2011, Neural Computation.

[4]  Eric Achten,et al.  Optimal Experimental Design for Diffusion Kurtosis Imaging , 2010, IEEE Transactions on Medical Imaging.

[5]  Sabine Van Huffel,et al.  Hierarchical non‐negative matrix factorization (hNMF): a tissue pattern differentiation method for glioblastoma multiforme diagnosis using MRSI , 2013, NMR in biomedicine.

[6]  Sanjeev Arora,et al.  Computing a nonnegative matrix factorization -- provably , 2011, STOC '12.

[7]  Yu-Jin Zhang,et al.  Nonnegative Matrix Factorization: A Comprehensive Review , 2013, IEEE Transactions on Knowledge and Data Engineering.

[8]  Jan Sijbers,et al.  Weighted linear least squares estimation of diffusion MRI parameters: Strengths, limitations, and pitfalls , 2013, NeuroImage.

[9]  Sabine Van Huffel,et al.  Hierarchical non-negative matrix factorization applied to in vivo 3T MRSI prostate data for automatic detection and visualization of tumours , 2013 .

[10]  Paulo J. G. Lisboa,et al.  Non-negative matrix factorisation methods for the spectral decomposition of MRS data from human brain tumours , 2012, BMC Bioinformatics.

[11]  Sabine Van Huffel,et al.  Frequency-selective quantitation of short-echo time 1H magnetic resonance spectra. , 2007, Journal of magnetic resonance.

[12]  Christos Boutsidis,et al.  SVD based initialization: A head start for nonnegative matrix factorization , 2008, Pattern Recognit..

[13]  Benoit Rivard,et al.  The Successive Projection Algorithm (SPA), an Algorithm with a Spatial Constraint for the Automatic Search of Endmembers in Hyperspectral Data , 2008, Sensors.

[14]  Hanli Qiao,et al.  New SVD based initialization strategy for non-negative matrix factorization , 2014, Pattern Recognit. Lett..

[15]  Guy Marchal,et al.  Multimodality image registration by maximization of mutual information , 1997, IEEE Transactions on Medical Imaging.

[16]  Nicolas Gillis,et al.  Fast and Robust Recursive Algorithmsfor Separable Nonnegative Matrix Factorization , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  Faculteit Ingenieurswetenschappen,et al.  QUANTIFICATION AND CLASSIFICATION OF MAGNETIC RESONANCE SPECTROSCOPIC DATA FOR BRAIN TUMOR DIAGNOSIS , 2008 .

[18]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[19]  Xue Xiao,et al.  A novel initialization method for nonnegative matrix factorization and its application in component recognition with three-dimensional fluorescence spectra. , 2012, Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy.

[20]  Sabine Van Huffel,et al.  Exploiting spatial information to estimate metabolite levels in two‐dimensional MRSI of heterogeneous brain lesions , 2011, NMR in biomedicine.

[21]  Michael W. Berry,et al.  Gene Tree Labeling Using Nonnegative Matrix Factorization on Biomedical Literature , 2008, Comput. Intell. Neurosci..

[22]  R M Weisskoff,et al.  Relative cerebral blood volume maps corrected for contrast agent extravasation significantly correlate with glioma tumor grade, whereas uncorrected maps do not. , 2006, AJNR. American journal of neuroradiology.

[23]  Chris Rorden,et al.  Improving Lesion-Symptom Mapping , 2007, Journal of Cognitive Neuroscience.

[24]  S. Bauer,et al.  A survey of MRI-based medical image analysis for brain tumor studies , 2013, Physics in medicine and biology.

[25]  Nicolas Gillis,et al.  Successive Nonnegative Projection Algorithm for Robust Nonnegative Blind Source Separation , 2013, SIAM J. Imaging Sci..

[26]  Sabine Van Huffel,et al.  Hierarchical non‐negative matrix factorization to characterize brain tumor heterogeneity using multi‐parametric MRI , 2015, NMR in biomedicine.

[27]  Brian B. Avants,et al.  The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) , 2015, IEEE Transactions on Medical Imaging.

[28]  Tomohiko Mizutani,et al.  Robustness analysis of preconditioned successive projection algorithm for general form of separable NMF problem , 2015, 1506.08387.

[29]  Reza Boostani,et al.  An Efficient Initialization Method for Nonnegative Matrix Factorization , 2011 .

[30]  Miroslav Zivanovic,et al.  ECG-EMG separation by using enhanced non-negative matrix factorization , 2014, 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[31]  C. D. Meyer,et al.  Initializations for the Nonnegative Matrix Factorization , 2006 .

[32]  Corrado Mencar,et al.  Subtractive clustering for seeding non-negative matrix factorizations , 2014, Inf. Sci..

[33]  Xinhe Xu,et al.  Facial expression recognition based on PCA and NMF , 2008, 2008 7th World Congress on Intelligent Control and Automation.

[34]  Stefan M. Wild,et al.  Improving non-negative matrix factorizations through structured initialization , 2004, Pattern Recognit..

[35]  Jie Yang,et al.  Initialization enhancer for non-negative matrix factorization , 2007, Eng. Appl. Artif. Intell..

[36]  Chih-Jen Lin,et al.  Projected Gradient Methods for Nonnegative Matrix Factorization , 2007, Neural Computation.

[37]  Frank Nielsen,et al.  On weighting clustering , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[38]  Chris H. Q. Ding,et al.  Convex and Semi-Nonnegative Matrix Factorizations , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.