Maximum Likelihood Estimation
Expectation-Maximization algorithms, or em algorithms for short, are iterative algorithms designed to solve maximum likelihood estimation problems. The general setting is that one observes a random sample \({Y }_{1},{Y }_{2},\ldots,{Y }_{n}\) of a random variable Y whose probability density function (pdf) \(f(\,\cdot \,\vert \,{x}_{o})\) with respect to some (known) dominating measure is known up to an unknown “parameter” \({x}_{o}\). The goal is to estimate \({x}_{o}\) and, one might add, to do it well. In this chapter that means to solve the maximum likelihood problem
and to solve it by means of em algorithms. The solution, assuming it exists and is unique, is called the maximum likelihood estimator of \({x}_{o}\). Here, the estimator is typically denoted by \(\widehat{x}\).
The notion of em algorithms was coined by [27], who unified various...
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References and Further Reading
Aronszajn N, Smith KT (1961) Theory of Bessel potentials. I. Ann Inst Fourier (Grenoble) 11:385–475, www.numdam.org
Atkinson KE (1969) The numerical solution of integral equations on the half line. SIAM J Numer Anal 6:375–397
Bardsley JM, Luttman A (2009) Total variation-penalized Poisson likelihood estimation for ill-posed problems. Adv Comput Math 31:35–39
Bertero M, Bocacci P, Desiderá G, Vicidomini G (2009) Image de-blurring with Poisson data: from cells to galaxies. Inverse Probl 25(123006):26
Browne J, De Pierro AR (1996) A row-action alternative to the EM algorithm for maximizing likelihoods in emission tomography. IEEE Trans Med Imag 15:687–699
Brune C, Sawatzky A, Burger M (2009) Bregman-EM-TV methods with application to optical nanoscopy, scale space and variational methods in computer vision, Lecture Notes in Computer Science 5567. Springer, Berlin, pp 235–246
Byrne CL (1993) Iterative image reconstruction algorithms based on cross-entropy minimization. IEEE Trans Image Process 2:96–103
Byrne CL (1996) Block-iterative methods for image reconstruction from projections. IEEE Trans Image Process 5:792–794
Byrne CL (1998) Accelerating the EMML algorithm and related iterative algorithms by rescaled block-iterative methods. IEEE Trans Image Process 7:792–794
Byrne CL (2001) Likelihood maximization for list-mode emission tomographic image reconstruction. IEEE Trans Med Imag 20:1084–1092
Byrne CL (2005) Choosing parameters in block-iterative or ordered subset reconstruction algorithms. IEEE Trans Image Process 14:321–327
Byrne CL (2005) Signal processing: a mathematical approach. AK Peters, Wellesley
Byrne CL (2008) Applied iterative methods. AK Peters, Wellesley
Byrne CL, Fiddy MA (1988) Images as power spectra; reconstruction as a Wiener filter approximation. Inverse Probl 4:399–409
Cao Yu, Eggermont PPB, Terebey S (1999) Cross Burg entropy maximization and its application to ringing suppression in image reconstruction. IEEE Trans Image Process 8:286–292
Censor Y, Eggermont PPB, Gordon D (1983) Strong under relaxation in Kaczmarz’s method for inconsistent systems. Numer Math 41:83–92
Censor Y, Lent AH (1987) Optimization of “log x” entropy over linear equality constraints. SIAM J Control Optim 25:921–933
Censor Y, Segman J (1987) On block-iterative entropy maximization. J Inform Optim Sci 8: 275–291
Censor Y, Zenios SA (1992) Proximal minimization algorithm with D-functions. J Optim Theory Appl 73:451–464
Cover TM (1984) An algorithm for maximizing expected log investment return. IEEE Trans Inform Theory 30:369–373
Crowther RA, DeRosier DJ, Klug A (1971) The reconstruction of three-dimensional structure from projections and its application to electron microscopy. Proc R Soc Lond A Math Phys Sci 317(3):19–340
Csiszár I (1975) I-divergence geometry of probability distributions and minimization problems. Ann Probab 3:146–158
Csiszár I, Tusnády G (1984) Information geometry and alternating minimization procedures. Stat Decisions 1(Supplement 1):205–237
Daley DJ, Vere-Jones D (2003) An introduction to the theory of point processes. Springer, New York
Darroch JN, Ratcliff D (1972) Generalized iterative scaling for log-linear models. Ann Math Stat 43:1470–1480
Daube-Witherspoon ME, Muehllehner G (1986) An iterative space reconstruction algorithm suitable for volume ECT. IEEE Trans Med Imag 5: 61–66
Dempster AP, Laird NM, Rubin DB (1977) Maximum likelihood from incomplete data via the EM algorithm. J R Stat Soc B 37:1–38
De Pierro AR (1987) On the convergence of the iterative image space reconstruction algorithm for volume ECT. IEEE Trans Med Imag 6: 174–175
De Pierro AR (1995) A modified expectation maximization algorithm for penalized likelihood estimation in emission tomography. IEEE Trans Med Imag 14:132–137
De Pierro A, Yamaguchi M (2001) Fast EM-like methods for maximum a posteriori estimates in emission tomography. Trans Med Imag 20: 280–288
Dey N, Blanc-Ferraud L, Zimmer Ch, Roux P, Kam Z, Olivo-Martin J-Ch, Zerubia J (2006) Richardson-Lucy algorithm with total variation regularization for 3D confocal microscope deconvolution. Microsc Res Tech 69:260–266
Duijster A, Scheunders P, De Backer S (2009) Wavelet-based EM algorithm for multispectral-image restoration. IEEE Trans Geoscience Remote Sensing 47:3892–3898
Eggermont PPB (1990) Multiplicative iterative algorithms for convex programming. Linear Algebra Appl 130:25–42
Eggermont PPB (1999) Nonlinear smoothing and the EM algorithm for positive integral equations of the first kind. Appl Math Optimiz 39: 75–91
Eggermont PPB, Herman GT, Lent AH (1981) Iterative algorithms for large partitioned linear systems with applications to image reconstruction. Linear Algebra Appl 40:37–67
Eggermont PPB, LaRiccia VN (1995) Smoothed maximum likelihood density estimation for inverse problems. Ann Stat 23:199–220
Eggermont PPB, LaRiccia VN (1997) Maximum penalized likelihood estimation and smoothed EM algorithms for positive integral equations of the first kind. Numer Funct Anal Optimiz 17:737–754
Eggermont PPB, LaRiccia VN (1998) On EM-like algorithms for minimum distance estimation. Manuscript, University of Delaware
Eggermont PPB, LaRiccia VN (2001) Maximum penalized likelihood estimation, I: Density estimation. Springer, New York
Elfving T (1980) On some methods for entropy maximization and matrix scaling. Linear Algebra Appl 34:321–339
Fessler JA, Ficaro EP, Clinthorne NH, Lange K (1997) Grouped coordinate ascent algorithms for penalized log-likelihood transmission image reconstruction. IEEE Trans Med Imag 16:166–175
Fessler JA, Hero AO (1995) Penalized maximum-likelihood image reconstruction using space-alternating generalized EM algorithms. IEEE Trans Image Process 4:1417–1429
Figueiredo MAT, Nowak RD (2003) An EM algorithm for wavelet-based image restoration. IEEE Trans Image Process 12:906–916
Frank J (2006) Three-dimensional electron microscopy of macromolecular assemblies, 2nd edn. Oxford University Press, New York
Geman S, Geman D (1984) Stochastic relaxation, Gibbs distributions and the Bayesian restoration of images. IEEE Trans Pattern Anal Mach Intell 6:721–741
Geman S, McClure DE (1985) Bayesian image analysis, an application to single photon emission tomography, Statistical Computing Section. Proc Am Stat Assoc 12–18
Good IJ (1971) A nonparametric roughness penalty for probability densities. Nature 229: 29–30
Gordon R, Bender R, Herman GT (1970) Algebraic reconstruction techniques (ART) for three-dimensional electron microscopy and X-ray photography. J Theor Biol 29:471–482
Green PJ (1990) Bayesian reconstructions from emission tomography data using a modified EM algorithm. IEEE Trans Med Imag 9:84–93
Guillaume M, Melon P, Réfrégier P (1998) Maximum-likelihood estimation of an astronomical image from a sequence at low photon levels. J Opt Soc Am A 15:2841–2848
Haltmeier M, Leitão A, Resmerita E (2009) On regularization methods of EM-Kaczmarz type. Inverse Probl 25(075008):17
Hanke M (1991) Accelerated Landweber iterations for the solution of ill-posed problems. Numer Math 60:341–373
Hartley HO (1958) Maximum likelihood estimation from incomplete data. Biometrics 14: 174–194
Hebert T, Leahy R (1989) A generalized EM algorithm for 3-D Bayesian reconstruction from Poisson data using Gibbs priors. IEEE Trans Med Imag 8:194–202
Herman GT (2009) Fundamentals of computerized tomography: image reconstruction from projections. Springer, New York
Herman GT, Meyer LB (1993) Algebraic reconstruction techniques can be made computationally efficient. IEEE Trans Med Imag 12: 600–609
Holte S, Schmidlin P, Lindén A, Rosenqvist G, Eriksson L (1990) Iterative image reconstruction for positron emission tomography: a study of convergence and quantitation problems. IEEE Trans Nuclear Sci 37:629–635
Horváth I, Bagoly Z, Balász LG, de Ugarte Postigo A, Veres P, Mészáros A (2010) Detailed classification of Swift’s Gamma-ray bursts. J Astrophys 713:552–557
Hudson HM, Larkin RS (1994) Accelerated image reconstruction using ordered subsets of projection data. IEEE Trans Med Imag 13:601–609
Kamphuis C, Beekman FJ, Viergever MA (1996) Evaluation of OS-EM vs. EM-ML for 1D, 2D and fully 3D SPECT reconstruction. IEEE Trans Nucl Sci 43:2018–2024
Kondor A (1983) Method of convergent weights – an iterative procedure for solving Fredholm’s integral equations of the first kind. Nucl Instrum Methods 216:177–181
Lange K (1990) Convergence of EM image reconstruction algorithms with Gibbs smoothing. IEEE Trans Med Imag 9:439–446
Lange K, Bahn M, Little R (1987) A theoretical study of some maximum likelihood algorithms for emission and transmission tomography. IEEE Trans Med Imag 6:106–114
Lange K, Carson R (1984) EM reconstruction algorithms for emission and transmission tomography. J Comput Assisted Tomography 8:306–316
Latham GA (1995) Existence of EMS solutions and a priori estimates. SIAM J Matrix Anal Appl 16:943–953
Levitan E, Chan M, Herman GT (1995) Image-modeling Gibbs priors. Graph Models Image Process 57:117–130
Lewitt RM, Muehllehner G (1986) Accelerated iterative reconstruction in PET and TOFPET. IEEE Trans Med Imag 5:16–22
Liu C, Rubin H (1994) The ECME algorithm: a simple extension of EM and ECM with faster monotone convergence. Biometrika 81:633–648
Llacer J, Veklerov E (1989) Feasible images and practical stopping rules for iterative algorithms in emission tomography. IEEE Trans Med Imag 8:186–193
Lucy LB (1974) An iterative technique for the rectification of observed distributions. Astronomical J 79:745–754
McLachlan GJ, Krishnan T (2008) The EM algorithm and its extensions. Wiley, Hoboken
Meidunas E (2001) Re-scaled block iterative expectation maximization maximum likelihood (RBI-EMML) abundance estimation and sub-pixel material identification in hyperspectral imagery. MS thesis, Department of Electrical Engineering, University of Massachusetts Lowell
Miller MI, Roysam B (1991) Bayesian image reconstruction for emission tomography incorporating Good’s roughness prior on massively parallel processors. Proc Natl Acad Sci USA 88:3223–3227
Mülthei HN, Schorr B (1987) On an iterative method for a class of integral equations of the first kind. Math Meth Appl Sci 9:137–168
Mülthei HN, Schorr B (1989) On properties of the iterative maximum likelihood reconstruction method. Math Meth Appl Sci 11:331–342
Nielsen SF (2006) The stochastic EM algorithm: estimation and asymptotic results. Bernoulli 6:457–489
Parra L, Barrett H (1998) List-mode likelihood: EM algorithm and image quality estimation demonstrated on 2-D PET. IEEE Trans Med Imag 17:228–235
Penczek P, Zhu J, Schroeder R, Frank J (1997) Three-dimensional reconstruction with contrast transfer function compensation. Scanning Microscopy 11:147–154
Redner RA, Walker HF (1984) Mixture densities, maximum likelihood and the EM algorithm. SIAM Rev 26:195–239
Resmerita E, Engl HW, Iusem AN (2007) The expectation-maximization algorithm for ill-posed integral equations: a convergence analysis. Inverse Probl 23:2575–2588
Richardson WH (1972) Bayesian based iterative method of image restoration. J Opt Soc Am 62:55–59
Rockmore A, Macovski A (1976) A maximum likelihood approach to emission image reconstruction from projections. IEEE Trans Nucl Sci 23:1428–1432
Scheres SHW, Valle M, Núñez R, Sorzano COS, Marabini R, Herman GT, Carazo J-M (2005) Maximum-likelihood multi-reference refinement for electron microscopy images. J Mol Biol 348:139–149
Scheres SHW, Gao HX, Valle M, Herman GT, Eggermont PPB, Frank J, Carazo J-M (2007a) Disentangling conformational states of macromolecules in 3D-EM through likelihood optimization. Nat Methods 4:27–29
Scheres SHW, Núñez-Ramírez R, Gómez-Llorente Y, San Martín C, Eggermont PPB, Carazo J-M (2007b) Modeling experimental image formation for likelihood-based classification of electron microscopy. Structure 15:1167–1177
Schmidlin P (1972) Iterative separation of tomographic scintigrams. Nuklearmedizin 11:1–16
Setzer S, Steidl G, Teuber T (2010) Deblurring Poissonian images by split Bregman techniques. J Vis Commun Image Repr 21:193–199
Shepp LA, Vardi Y (1982) Maximum likelihood reconstruction in emission tomography. IEEE Trans Med Imag 1:113–122
Sigworth FJ (1998) A maximum-likelihood approach to single-particle image refinement. J Struct Biol 122:328–339
Silverman BW, Jones MC, Wilson JD, Nychka DW (1990) A smoothed EM algorithm approach to indirect estimation problems, with particular reference to stereology and emission tomography (with discussion). J R Stat Soc B 52:271–324
Sun Y, Walker JG (2008) Maximum likelihood data inversion for photon correlation spectroscopy. Meas Sci Technol 19(115302):8
Tanaka E, Kudo H (2010) Optimal relaxation parameters of DRAMA (dynamic RAMLA) aiming at one-pass image reconstruction for 3D-PET. Phys Med Biol 55:2917–2939
Tarasko MZ (1969) On a method for solution of the linear system with stochastic matrices (in Russian), Report Physics and Energetics Institute, Obninsk PEI-156
Trummer MR (1984) A note on the ART of relaxation. Computing 33:349–352
van der Sluis A, van der Vorst HA (1990) SIRT- and CG-type methods for the iterative solution of sparse linear least-squares problems. Linear algebra in image reconstruction from projections. Linear Algebra Appl 130: 257–303
Vardi Y, Shepp LA, Kaufman L (1985) A statistical model for positron emission tomography (with discussion). J Am Stat Assoc 80:8–38
Wernick M, Aarsvold J (2004) Emission tomography: the fundamentals of PET and SPECT. Elsevier Academic Press, San Diego
Wu CFJ (1983) On the convergence properties of the EM algorithm. Ann Stat 11:95–103
Yu S, Latham GA, Anderssen RS (1994) Stabilizing properties of maximum penalized likelihood estimation for additive Poisson regression. Inverse Probl 10:1199–1209
Yuan Jianhua, Yu Jun (2007) Median-prior tomography reconstruction combined with nonlinear anisotropic diffusion filtering. J Opt Soc Am A 24: 1026–1033
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer Science+Business Media, LLC
About this entry
Cite this entry
Byrne, C., Eggermont, P.P.B. (2011). EM Algorithms. In: Scherzer, O. (eds) Handbook of Mathematical Methods in Imaging. Springer, New York, NY. https://doi.org/10.1007/978-0-387-92920-0_8
Download citation
DOI: https://doi.org/10.1007/978-0-387-92920-0_8
Publisher Name: Springer, New York, NY
Print ISBN: 978-0-387-92919-4
Online ISBN: 978-0-387-92920-0
eBook Packages: Mathematics and StatisticsReference Module Computer Science and Engineering