Using deconvolution to improve PET spatial resolution in OSEM iterative reconstruction

Institute of Bioimaging and Molecular Physiology (IBFM)-CNR, University of Milano-Bicocca, San Raffaele Scientific Institute, Milan, Italy.
Methods of Information in Medicine (Impact Factor: 2.25). 02/2007; 46(2):231-5.
Source: PubMed


A novel approach to the PET image reconstruction is presented, based on the inclusion of image deconvolution during conventional OSEM reconstruction. Deconvolution is here used to provide a recovered PET image to be included as "a priori" information to guide OSEM toward an improved solution.
Deconvolution was implemented using the Lucy-Richardson (LR) algorithm: Two different deconvolution schemes were tested, modifying the conventional OSEM iterative formulation: 1) We built a regularizing penalty function on the recovered PET image obtained by deconvolution and included it in the OSEM iteration. 2) After each conventional global OSEM iteration, we deconvolved the resulting PET image and used this "recovered" version as the initialization image for the next OSEM iteration. Tests were performed on both simulated and acquired data.
Compared to the conventional OSEM, both these strategies, applied to simulated and acquired data, showed an improvement in image spatial resolution with better behavior in the second case. In this way, small lesions, present on data, could be better discriminated in terms of contrast.
Application of this approach to both simulated and acquired data suggests its efficacy in obtaining PET images of enhanced quality.

Download full-text


Available from: Giovanna Rizzo, Jul 02, 2014
1 Follower
51 Reads
  • Source
    • "Finally, a recent work of Rahmim et al. (2008) strictly applied the idea of equation (4), putting the positron range correction in image space and the other corrections in sinogram space. Other works and applications on clinical data include Rahmim, Cheng and Sossi (2005), Rizzo et al. (2007), Varrone et al. (2009), Mourik et al. (2010) and Hoetjes et al. (2010). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Partial volume effect is an important source of bias in PET images that can be lowered by accounting for the point spread function (PSF) of the scanner. We measured such a PSF in various points of a clinical PET scanner and modelled it as a product of matrices acting in image space, taking the asymmetrical, shift-varying and non-Gaussian character of the PSF into account (AMP modelling), and we integrated this accurate image space modelling into a conventional list-mode OSEM algorithm (EM-AMP reconstruction). We showed on the one hand that when a sufficiently high number of iterations are considered, the AMP modelling lead to better recovery coefficients at reduced background noise compared to reconstruction where no or only partial resolution modelling is performed, and on the other hand that for a small number of iterations, a Gaussian modelling gave the best recovery coefficients. Moreover, we have demonstrated that a deconvolution based on the AMP system response model leads to the same recovery coefficients as the corresponding EM-AMP reconstruction, but at the expense of an increased background noise.
    Physics in Medicine and Biology 09/2010; 55(17):5045-66. DOI:10.1088/0031-9155/55/17/011 · 2.76 Impact Factor
  • Source
    • "In an even more recent paper, deconvolution was performed prior to each subsequent iteration during ordered sub-sets expectation maximization (OSEM) reconstruction (Rizzo et al 2007). The second category includes methods, which apply a resolution correction to the PET images post-reconstruction. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Correcting positron emission tomography (PET) images for the partial volume effect (PVE) due to the limited resolution of PET has been a long-standing challenge. Various approaches including incorporation of the system response function in the reconstruction have been previously tested. We present a post-reconstruction PVE correction based on iterative deconvolution using a 3D maximum likelihood expectation-maximization (MLEM) algorithm. To achieve convergence we used a one step late (OSL) regularization procedure based on the assumption of local monotonic behavior of the PET signal following Alenius et al. This technique was further modified to selectively control variance depending on the local topology of the PET image. No prior 'anatomic' information is needed in this approach. An estimate of the noise properties of the image is used instead. The procedure was tested for symmetric and isotropic deconvolution functions with Gaussian shape and full width at half-maximum (FWHM) ranging from 6.31 mm to infinity. The method was applied to simulated and experimental scans of the NEMA NU 2 image quality phantom with the GE Discovery LS PET/CT scanner. The phantom contained uniform activity spheres with diameters ranging from 1 cm to 3.7 cm within uniform background. The optimal sphere activity to variance ratio was obtained when the deconvolution function was replaced by a step function few voxels wide. In this case, the deconvolution method converged in approximately 3-5 iterations for most points on both the simulated and experimental images. For the 1 cm diameter sphere, the contrast recovery improved from 12% to 36% in the simulated and from 21% to 55% in the experimental data. Recovery coefficients between 80% and 120% were obtained for all larger spheres, except for the 13 mm diameter sphere in the simulated scan (68%). No increase in variance was observed except for a few voxels neighboring strong activity gradients and inside the largest spheres. Testing the method for patient images increased the visibility of small lesions in non-uniform background and preserved the overall image quality. Regularized iterative deconvolution with variance control based on the local properties of the PET image and on estimated image noise is a promising approach for partial volume effect corrections in PET.
    Physics in Medicine and Biology 06/2008; 53(10):2577-91. DOI:10.1088/0031-9155/53/10/009 · 2.76 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Super-resolution (SR) techniques are used in PET imaging to generate a high-resolution image by combining multiple low-resolution images that have been acquired from different points of view (POV). In this article, the authors propose a novel implementation of the SR technique whereby the required multiple low-resolution images are generated by shifting the reconstruction pixel grid during the image reconstruction process rather than being acquired from different POVs. The objective of this article is to compare the performances of the two SR implementations using theoretical and experimental studies. A mathematical framework is first provided to support the hypothesis that the two SR implementations have similar performance in current PET/CT scanners that use block detectors. Based on this framework, a simulation study, a point source study, and a NEMA/IEC phantom study were conducted to compare the performance of these two SR implementations with respect to contrast, resolution, noise, and SNR. For reference purposes, a comparison with a native reconstruction (NR) image using a high-resolution pixel grid was also performed. The mathematical framework showed that the two SR implementations are expected to achieve similar contrast and resolution but different noise contents. These results were confirmed by the simulation and experimental studies. The simulation study showed that the two SR implementations have an average contrast difference of 2.3%, while the point source study showed that their average differences in contrast and resolution were 0.5% and 1.2%, respectively. Comparisons between the SR and NR images for the point source study showed that the NR image exhibited averages of 30% and 8% lower contrast and resolution, respectively. The NEMA/IEC phantom study showed that the three images (two SR and NR) exhibited different noise structures. The SNR of the new SR implementation was, on average, 21.5% lower than the original implementation largely due to an increase in background noise, while the NR image had averages of 18.5% and 8% lower SNR and contrast, respectively, versus the two SR images. The new SR implementation can potentially replace the original SR approach in current PET scanners that use block detectors while maintaining similar contrast and resolution, but at a relatively lower SNR. A major advantage of the new SR implementation is its shorter overall scan duration which results in an increase in scanner throughput and a reduction in patient motion.
    Medical Physics 05/2009; 36(4):1370-83. DOI:10.1118/1.3090890 · 2.64 Impact Factor
Show more