Abstract
Magnitude magnetic resonance (MR) images are noise-contaminated measurements of the true signal, and it is important to assess the noise in many applications. A recently introduced approach models the magnitude MR datum at each voxel in terms of a mixture of up to one Rayleigh and an a priori unspecified number of Rice components, all with a common noise parameter. The Expectation-Maximization (EM) algorithm was developed for parameter estimation, with the mixing component membership of each voxel as the missing observation. This paper revisits the EM algorithm by introducing more missing observations into the estimation problem such that the complete (observed and missing parts) dataset can be modeled in terms of a regular exponential family. Both the EM algorithm and variance estimation are then fairly straightforward without any need for potentially unstable numerical optimization methods. Compared to local neighborhood- and wavelet-based noise-parameter estimation methods, the new EM-based approach is seen to perform well not only in simulation experiments but also on physical phantom and clinical imaging data.
Similar content being viewed by others
Notes
Note that some authors use the spelling “Rician”: we follow others in using “Ricean” since the adjective is derived from the name of S. O. Rice.
References
Abramowitz, M. and Stegun, I.E. (1972). Handbook of Mathematical Functions. Dover Publications, New York.
Ahmed, O.A. (2005). New denoising scheme for magnetic resonance spectroscopy signals. IEEE Trans. Med. Imag., 24, 809–816.
Aja-Fernández, S., Tristán-Vega, A. and Alberola-Lòpez, C. (2009). Noise estimation in single- and multiple-coil magnetic resonance data based on statistical models. Magn. Reson. Imaging, 27, 1397–1409.
Altman, M., Gill, J. and Mcdonald, M. (2003). Numerical Issues in Statistical Computing for the Social Scientist. Wiley-Interscience, New York.
Bammer, R., Skare, S., Newbould, R., Liu, C., Thijs, V., Ropele, S., Clayton, D.B., Krueger, G., Moseley, M.E. and Glover, G.H. (2005). Foundations of advanced magnetic resonance imaging. NeuroRx, 2, 167–196.
Biernacki, C., Celeux, G. and Gold, E.M. (2000). Assessing a mixture model for clustering with the integrated completed likelihood. IEEE Trans. Pattern Anal. Mach. Intell., 22, 719–725.
Biernacki, C., Celeux, G. and Govaert, G. (2003). Choosing starting values for the EM algorithm for getting the highest likelihood in multivariate Gaussian mixture models. Comput. Statist. Data Anal., 413, 561–575.
Brown, T.R., Kincaid, B.M. and Ugurbil, K. (1982). NMR chemical shift imaging in three dimensions. Proc. Natl. Acad. Sci. USA, 79, 3523–3526.
Brummer, M.E., Mersereau, R.M., Eisner, R.L. and Lewine, R.R.J. (1993). Automatic detection of brain contours in MRI data sets. IEEE Trans. Med. Imag., 12.
Byrd, R.H., Lu, P., Nocedal, J. and Zhu, C. (1995). A limited memory algorithm for bound constrained optimization. SIAM J. Sci. Comput., 16, 1190–1208.
Cocosco, C., Kollokian, V., Kwan, R. and Evans, A. (1997). Brainweb: online interface to a 3d MRI simulated brain database. NeuroImage, 5.
Coupè, P., Manjn, J.V., Gedamu, E., Arnold, D., Robles, M. and Collins, D.L. (2010). Robust Rician noise estimation for MR images. Med. Image Anal., 14, 483–493.
Daubechies, I. (1992). Ten Lectures on Wavelets. In CBMS-NSF Regional Conference Series in Applied Mathematics. SIAM, Philadelphia.
Dempster, A., Laird, N. and Rubin, D. (1977). Maximum likelihood from incomplete data via the EM algorithm. J. R. Stat. Soc. Ser. B, 39, 1–38.
Donoho, D.L. (1995). De-noising by soft-thresholding. IEEE Trans. Inform. Theory, 41, 613–627.
Donoho, D.L. and Johnstone, I.M. (1994). Ideal spatial adaptation by wavelet shrinkage. Biometrika, 81, 425–455.
Fraley, C. and Raftery, A.E. (2002). Model-based clustering, discriminant analysis, and density estimation. J. Amer. Statist. Assoc., 97, 611–631.
Glad, I.K. and Sebastiani, G. (1995). A Bayesian approach to Synthetic Magnetic Resonance Imaging. Biometrika, 82, 237–250.
Hartigan, J.A. and Wong, M.A. (1979). A k-means clustering algorithm. Applied Statistics, 28, 100–108.
Hennessy, M.J. (2000). A three-dimensional physical model of MRI noise based on current noise sources in a conductor. J. Magn. Reson., 147, 153–169.
Hinshaw, W.S. and Lent, A.H. (1983). An introduction to NMR imaging: From the Bloch equation to the imaging equation. Proc. IEEE, 71.
Keribin, C. (2000). Consistent estimation of the order of finite mixture models. Sankhyā A, 62, 49–66.
Koay, C.G. and Basser, P.J. (2006). Analytically exact correction scheme for signal extraction from noisy magnitude MR signals. J. Magn. Reson., 179, 317–322.
Ljunggren, S. (1983). A simple graphical representation of fourier-based imaging methods. J. Magn. Reson., 54, 338–343.
Louis, T. (1982). Finding the observed information matrix when using the EM algorithm. J. R. Stat. Soc. Ser. B, 44, 226–233.
Macqueen, J. (1967). Some methods of classification and analysis of multivariate observations. In Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, pp. 281–297.
Maitra, R. (2009). Initializing partition-optimization algorithms. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 6, 144–157.
Maitra, R. and Faden, D. (2009). Noise estimation in magnitude MR datasets. IEEE Trans. Med. Imag., 28, 1615–1622.
Maitra, R. and Melnykov, V. (2010). Simulating data to study performance of finite mixture modeling and clustering algorithms. J. Comput. Graph. Statist.
Maitra, R. and Riddles, J.J. (2010). Synthetic Magnetic Resonance Imaging revisited. IEEE Trans. Med. Imag., 29, 895–902.
Mardia, K.V. and Jupp, P.E. (2000). Directional Statistics. Wiley, New York.
McLachlan, G. and Krishnan, T. (2008). The EM Algorithm and Extensions, second edition. Wiley, New York.
McLachlan, G. and Peel, D. (2000). Finite Mixture Models. John Wiley and Sons, Inc., New York.
McVeigh, E.R., Henkelman, R.M. and Bronskill, M.J. (1985). Noise and filtration in Magnetic Resonance Imaging. Med. Phys., 12, 586–591.
Pasquale, F.D., Barone, P., Sebastiani, G. and Stander, J. (2004). Bayesian analysis of dynamic magnetic resonance breast images. J. R. Stat. Soc. Ser. C, 53, 475–493.
Rajan, J., Poot, D., Juntu, J. and Sijbers, J. (2010). Noise measurement from magnitude MRI using local estimates of variance and skewness. Phys. Med. Biol., 55, N441–449.
Rice, S.O. (1944). Mathematical analysis of random noise. Bell Syst. Tech. J., 23, 282.
Rice, S.O. (1945). Mathematical analysis of random noise. Bell Syst. Tech. J., 24, 46–156.
Rohdea, G.K., Barnettc, A.S., Bassera, P.J. and Pierpaoli, C. (2005). Estimating intensity variance due to noise in registered images: applications to diffusion tensor MRI. NeuroImage, 26, 673–684.
Schwarz, G. (1978). Estimating the dimension of a model. Ann. Statist., 6, 461–464.
Sijbers, J. (1998). Signal and Noise Estimation from Magnetic Resonance Images. PhD thesis, University of Antwerp.
Sijbers, J. and den Dekker, A.J. (2004). Maximum likelihood estimation of signal amplitude and noise variance from MR data. Magn. Reson. Med., 51, 586–594.
Sijbers, J., den Dekker, A.J., Van Audekerke, J., Verhoye, M. and Van Dyck, D. (1998). Estimation of the noise in magnitude MR images. Magn. Reson. Imaging, 16, 87–90.
Sijbers, J., Poot, D., den Dekker, A.J. and Pintjens, W. (2007). Automatic estimation of the noise variance from the histogram of a magnetic resonance image. Phys. Med. Biol., 52, 1335–1348.
Smith, R.C. and Lange, R.C. (2000). Understanding Magnetic Resonance Imaging. CRC Press LLC.
Tweig, D.B. (1983). The k-trajectory formulation of the NMR imaging process with applications in analysis and synthesis of imaging methods. Med. Phys., 10, 610–21.
Wang, T. and Lei, T. (1994). Statistical analysis of MR imaging and its application in image modeling. In Proceedings of the IEEE International Conference on Image Processing and Neural Networks, vol. 1, pp. 866–870.
Weishaupt, D., Köchli, V.D. and Marincek, B. (2003). How Does MRI Work? Springer–Verlag, New York.
Wilde, J.P.D., Lunt, J. and Straughan, K. (1997). Information in magnetic resonance images: evaluation of signal, noise and contrast. Med. Biol. Eng. Comput., 35, 259–265.
Zhang, Y., Brady, M. and Smith, S. (2001). Segmentation of brain MR images through a hidden Markov random field model and the expectation maximization algorithm. IEEE Trans. Med. Imag., 20, 45–47.
Zhu, C., Byrd, R.H., Lu, P. and Nocedal, J. (1994). L-BFGS-B – Fortran subroutines for large-scale bound constrained optimization. Technical report, Northwestern University.
Acknowledgement
I thank R. P. Gullapalli and S. R. Roys for the phantom and clinical datasets and to an Associate Editor and two reviewers whose helpful and insightful comments on an earlier version of this manuscript greatly improved its content. I also acknowledge partial support by the National Science Foundation under its NSF CAREER DMS-0437555 award.
Author information
Authors and Affiliations
Corresponding author
Appendix: Conditional distribution of \(\boldsymbol{\gamma_i}\) given \(\boldsymbol{R_i}\)
Appendix: Conditional distribution of \(\boldsymbol{\gamma_i}\) given \(\boldsymbol{R_i}\)
We first state and prove the following
Theorem 1
Let \(\boldsymbol{W}_i\) be a realization from the one-trial J -class multinomial distribution with class probability vector \(\boldsymbol{\pi}\) . Conditional on W i,j = 1 for some j = 1,2,...,J, let \(U_i\sim N(\nu_j, \sigma^2)\) be independent of \(V_i\sim N(0, \sigma^2)\) . Write (U i , V i ) in its polar form i.e. (U i , V i ) = (R i cosγ i , R i sinγ i ). Then the conditional distribution of γ i given R i and the event W i,j = 1 is \(\mathcal M(0, {R_i\nu_j}/{\sigma^2})\) , i.e. it is von-Mises-distributed with mean angular direction 0 and concentration parameter R i ν j /σ 2 . Consequently, \({\textrm{I\!E}}(\cos\gamma_i\mid R_i, W_{i,j} = 1) = {{\textrm{I}}_1({R_i\nu_j}/{\sigma^2} )}/{{\textrm{I}}_0({R_i\nu_j}/{\sigma^2})}\) .
Proof
From the characterization of the Rice distribution in Rice (1944, 1945), we know that \(R_i\sim \varrho(x;\sigma,\nu_j)\) given that W i,j = 1. From standard results on probability distributions of transformation of variables, we get that the conditional joint density of (R i ,γ i ) given that W i,j = 1 is \(f_{R_i,\gamma_i \mid W_{i,j}=1}(x, \gamma) = x/{2\pi\sigma^2}\exp{[-(x^2-2x\nu_j\cos\gamma_i + \nu_j^2)/2\sigma^2]}\). Thus \(f(\gamma_i | R_i=x, W_{i,j}=1) = \exp{[{x\nu_j}/{\sigma^2}\cos\gamma_i]} / {2\pi{\textrm{I}}_0({x\nu_j}/{\sigma^2})} \) for 0 < γ i < 2π, which is the density of \(\mathcal M(0, {x\nu_j}/{\sigma^2})\). From results in Mardia and Jupp (2000) on the expectation of the cosine of a von-Mises-distributed random variable, we get that \({\textrm{I\!E}}(\cos\gamma_i\mid R_i, W_{i,j} = 1) = {{\textrm{I}}_1({R_i\nu_j}/{\sigma^2} )}/{{\textrm{I}}_0({R_i\nu_j}/{\sigma^2})}\). This proves Theorem 1.□
Rights and permissions
About this article
Cite this article
Maitra, R. On the Expectation-Maximization algorithm for Rice-Rayleigh mixtures with application to noise parameter estimation in magnitude MR datasets. Sankhya B 75, 293–318 (2013). https://doi.org/10.1007/s13571-012-0055-y
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13571-012-0055-y
Keywords and phrases
- Bayes Information Criterion
- Integrated Completed Likelihood
- local skewness
- mixture model
- Rayleigh density
- Rice density
- robust noise estimation
- wavelets