Document Detail


Ordered subsets Bayesian tomographic reconstruction using 2-D smoothing splines as priors.
MedLine Citation:
PMID:  12850295     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
The ordered subsets expectation maximization (OS-EM) algorithm has enjoyed considerable interest for accelerating the well-known EM algorithm for emission tomography. The OS principle has also been applied to several regularized EM algorithms, such as nonquadratic convex minimization-based maximum a posteriori (MAP) algorithms. However, most of these methods have not been as practical as OS-EM due to their complex optimization methods and difficulties in hyperparameter estimation. We note here that, by relaxing the requirement of imposing sharp edges and using instead useful quadratic spline priors, solutions are much easier to compute, and hyperparameter calculation becomes less of a problem. In this work, we use two-dimensional smoothing splines as priors and apply a method of iterated conditional modes for the optimization. In this case, step sizes or line-search algorithms necessary for gradient-based descent methods are avoided. We also accelerate the resulting algorithm using the OS approach and propose a principled way of scaling smoothing parameters to retain the strength of smoothing for different subset numbers. Our experimental results show that the OS approach applied to our quadratic MAP algorithms provides a considerable acceleration while retaining the advantages of quadratic spline priors.
Authors:
Soo-Jin Lee
Related Documents :
15295985 - Experimental verification of an interpolation algorithm for improved estimates of anima...
20377455 - Lifting prediction to alignment of rna pseudoknots.
20964545 - Modeling multivariate time series on manifolds with skew radial basis functions.
16177695 - Use of multiple data streams to conduct bayesian biologic surveillance.
25194685 - Toward dna-based facial composites: preliminary results and validation.
21991305 - Selecting indicator portfolios for marine species and food webs: a puget sound case study.
Publication Detail:
Type:  Journal Article; Research Support, Non-U.S. Gov't    
Journal Detail:
Title:  Computer methods and programs in biomedicine     Volume:  72     ISSN:  0169-2607     ISO Abbreviation:  Comput Methods Programs Biomed     Publication Date:  2003 Sep 
Date Detail:
Created Date:  2003-07-09     Completed Date:  2003-12-08     Revised Date:  2006-11-15    
Medline Journal Info:
Nlm Unique ID:  8506513     Medline TA:  Comput Methods Programs Biomed     Country:  Ireland    
Other Details:
Languages:  eng     Pagination:  27-42     Citation Subset:  IM    
Affiliation:
Department of Electronic Engineering, Paichai University, 439-6 Doma 2-Dong, Seo-Ku, 302-735 Taejon, South Korea. sjlee@pcu.ac.kr
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Algorithms
Bayes Theorem
Brain / radionuclide imaging
Computer Simulation
Humans
Phantoms, Imaging
Tomography, Emission-Computed / statistics & numerical data*

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  'Minimum average risk' as a new peak-detection algorithm applied to myofibrillar dynamics.
Next Document:  Lossy compression of eye movement and auditory brainstem response signals.