Analysis of Penalized Likelihood Image Reconstruction for Dynamic PET Quantification

Quantification of tracer kinetics using dynamic positron emission tomography (PET) provides important information for understanding the physiological and biochemical processes in humans and animals. A common procedure is to reconstruct a sequence of dynamic images first, and then apply kinetic analysis to the time activity curve of a region of interest derived from the reconstructed images. Obviously, the choice of image reconstruction method and its parameters affect the accuracy of the time activity curve and hence the estimated kinetic parameters. This paper analyzes the effects of penalized likelihood image reconstruction on tracer kinetic parameter estimation. Approximate theoretical expressions are derived to study the bias, variance, and ensemble mean squared error of the estimated kinetic parameters. Computer simulations show that these formulae predict correctly the changes of these statistics as functions of the regularization parameter. It is found that the choice of the regularization parameter has a significant impact on kinetic parameter estimation, indicating proper selection of image reconstruction parameters is important for dynamic PET. A practical method has been developed to use the theoretical formulae to guide the selection of the regularization parameter in dynamic PET image reconstruction.

[1]  R. Huesman,et al.  An investigation into the effect of input function shape and image acquisition interval on estimates of washin for dynamic cardiac SPECT. , 1997, Physics in medicine and biology.

[2]  Raymond F. Muzic,et al.  Optimal experiment design for PET quantification of receptor concentration , 1996, IEEE Trans. Medical Imaging.

[3]  Chin-Tu Chen,et al.  Image Restoration Using Gibbs Priors: Boundary Modeling, Treatment of Blurring, and Selection of Hyperparameter , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Zhenyu Zhou,et al.  Approximate maximum likelihood hyperparameter estimation for Gibbs priors , 1995, Proceedings., International Conference on Image Processing.

[5]  E. DiBella,et al.  Analytical propagation of errors in dynamic SPECT: estimators, degrading factors, bias and noise. , 1999, Physics in medicine and biology.

[6]  Jeffrey A. Fessler,et al.  Analysis of observer performance in known-location tasks for tomographic image reconstruction , 2006, IEEE Transactions on Medical Imaging.

[7]  Ronald H. Huesman,et al.  Theoretical study of penalized-likelihood image reconstruction for region of interest quantification , 2006, IEEE Transactions on Medical Imaging.

[8]  Raymond F Muzic,et al.  Evaluation of objective functions for estimation of kinetic parameters. , 2006, Medical physics.

[9]  Jeffrey A. Fessler.  Mean and variance of implicitly defined biased estimators (such as penalized maximum likelihood): applications to tomography , 1996, IEEE Trans. Image Process..

[10]  Alan C. Evans,et al.  Minimisation of parameter estimation errors in dynamic PET: choice of scanning schedules. , 1989, Physics in medicine and biology.

[11]  D Feng,et al.  A computer simulation study on the input function sampling schedules in tracer kinetic modeling with positron emission tomography (PET). , 1994, Computer methods and programs in biomedicine.

[12]  J. Delforge,et al.  Identifiability analysis and parameter identification of an in vivo ligand-receptor model from PET data , 1990, IEEE Transactions on Biomedical Engineering.

[13]  Jinyi Qi,et al.  Penalized maximum-likelihood image reconstruction for lesion detection. , 2006, Physics in medicine and biology.

[14]  R H Huesman,et al.  Dynamic PET Data Analysis , 1986, Journal of computer assisted tomography.

[15]  Andrei V. Gribok,et al.  Information complexity-based regularization parameter selection for solution of ill conditioned inverse problems , 2002 .

[16]  G. Hutchins,et al.  Sampling requirements for dynamic cardiac PET studies using image-derived input functions. , 1993, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[17]  David Dagan Feng,et al.  Optimal image sampling schedule: a new effective way to reduce dynamic image storage space and functional image processing time , 1996, IEEE Trans. Medical Imaging.

[18]  G. Gindi,et al.  Noise analysis of MAP-EM algorithms for emission tomography. , 1997, Physics in medicine and biology.

[19]  Richard M. Leahy,et al.  Resolution and noise properties of MAP reconstruction for fully 3-D PET , 2000, IEEE Transactions on Medical Imaging.

[20]  Wan-Chi Siu,et al.  A technique for extracting physiological parameters and the required input function simultaneously from PET image measurements: theory and simulation study , 1997, IEEE Transactions on Information Technology in Biomedicine.

[21]  J C Mazziotta,et al.  Methods for improving quantitation of putamen uptake constant of FDOPA in PET studies. , 1993, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[22]  R. Huesman,et al.  Consequences of using a simplified kinetic model for dynamic PET data. , 1997, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[23]  S. Eberl,et al.  Effect of reconstruction and filtering on kinetic parameter estimation bias and reliability for dynamic SPECT: a simulation study , 2005, IEEE Transactions on Nuclear Science.

[24]  Alfred O. Hero,et al.  Exploring estimator bias-variance tradeoffs using the uniform CR bound , 1996, IEEE Trans. Signal Process..

[25]  Gene Gindi,et al.  Fast LROC analysis of Bayesian reconstructed emission tomographic images using model observers. , 2005, Physics in medicine and biology.

[26]  S. Eberl,et al.  Optimized acquisition time and image sampling for dynamic SPECT of Tl-201 , 1998, IEEE Transactions on Medical Imaging.

[27]  Jinyi Qi,et al.  Effect of errors in the system matrix on maximum a posteriori image reconstruction , 2005, Physics in medicine and biology.

[28]  J. Fessler,et al.  Spatial resolution properties of penalized-likelihood image reconstruction: space-invariant tomographs , 1996, 5th IEEE EMBS International Summer School on Biomedical Imaging, 2002..

[29]  David Dagan Feng,et al.  Optimal image sampling schedule for both image-derived input and output functions in PET cardiac studies , 2000, IEEE Transactions on Medical Imaging.

[30]  Finbarr O'Sullivan,et al.  Use of ridge regression for improved estimation of kinetic constants from PET data , 1999, IEEE Transactions on Medical Imaging.

[31]  Jinyi Qi,et al.  A unified noise analysis for iterative image estimation. , 2003, Physics in medicine and biology.

[32]  S. Huang,et al.  The effects of measurement errors in the plasma radioactivity curve on parameter estimation in positron emission tomography. , 1991, Physics in medicine and biology.

[33]  B. Tsui,et al.  Noise properties of the EM algorithm: II. Monte Carlo simulations. , 1994, Physics in medicine and biology.

[34]  Richard M. Leahy,et al.  A theoretical study of the contrast recovery and variance of MAP reconstructions from PET data , 1999, IEEE Transactions on Medical Imaging.

[35]  J Delforge,et al.  Experimental design optimisation: theory and application to estimation of receptor model parameters using dynamic positron emission tomography. , 1989, Physics in medicine and biology.

[36]  M E Phelps,et al.  An evaluation of Bayesian regression for estimating cerebral oxygen utilization with oxygen-15 and dynamic PET. , 1988, IEEE transactions on medical imaging.

[37]  H. V. Trees,et al.  Exploring Estimator BiasVariance Tradeoffs Using the Uniform CR Bound , 2007 .

[38]  Zheru Chi,et al.  Tracer kinetic modeling of /sup 11/C-acetate applied in the liver with positron emission tomography , 2004, IEEE Transactions on Medical Imaging.

[39]  Per Christian Hansen,et al.  Analysis of Discrete Ill-Posed Problems by Means of the L-Curve , 1992, SIAM Rev..

[40]  Jeffrey A. Fessler,et al.  Covariance of kinetic parameter estimators based on time activity curve reconstructions: preliminary study on 1D dynamic imaging , 2004, 2004 2nd IEEE International Symposium on Biomedical Imaging: Nano to Macro (IEEE Cat No. 04EX821).

[41]  M. Graham.  Physiologic smoothing of blood time-activity curves for PET data analysis. , 1997, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[42]  Ken D. Sauer,et al.  ML parameter estimation for Markov random fields with applications to Bayesian tomography , 1998, IEEE Trans. Image Process..

[43]  D. Mankoff,et al.  Kinetic analysis of 2-[carbon-11]thymidine PET imaging studies: compartmental model and mathematical analysis. , 1998, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[44]  Donald W. Wilson,et al.  Noise properties of the EM algorithm. I. Theory , 1994 .

[45]  Charles L. Byrne,et al.  Noise characterization of block-iterative reconstruction algorithms. I. Theory , 2000, IEEE Transactions on Medical Imaging.