Yves D’Asseler

Ghent University, Gent, VLG, Belgium

Are you Yves D’Asseler?

Claim your profile

Publications (9)3.43 Total impact

  • [show abstract] [hide abstract]
    ABSTRACT: Compared to classical parallel hole (PH) collimated gamma cameras, the geometry of a rotating slat (RS) collimated detector implies a better sensitivity versus resolution trade-off. Nevertheless, the expected improvement in image quality from this device is degraded due to the assumptions made during the image reconstruction of planar images. This study shows that an accurate model of the imaging process for iterative reconstruction is necessary in order to improve planar image quality. To this end, a Monte Carlo based system matrix was derived from a simple simulation by exploiting the rotational and translational symmetries present in this acquisition set-up. It was found that this system matrix improves the contrast and noise characteristics of reconstructed images compared to a reconstruction with no modelling. As a result, the RS device is able to yield improved quality for planar images compared to a PH camera.
    Nuclear Instruments and Methods in Physics Research Section A Accelerators Spectrometers Detectors and Associated Equipment 01/2007; 571(1):274-277. · 1.14 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: The Crystal Clear Collaboration has designed and built a family of high resolution small animal PET scanners. These were designed to be used in research laboratories and provide maximum modularity and flexibility. The source code of the data acquisition and reconstruction software is freely available to the users. The design is based on the use of the Hamamatsu R7600-M64 multi-anode photomultiplier tube and an LSO/LuYAP phoswich matrix with one-to-one coupling between the crystals and the photo-detector. A complete system has 80 PMT tubes in four rings with a minimum inner diameter of 137 mm and an axial field of view of 110 mm. The detectors are rotating over 360 degrees so that partially filled ring geometries can be used. This greatly simplifies the combination of PET with other imaging modalities. Single gamma interactions are recorded in list mode format and coincidences are found by software.
    03/2006: pages 149-164;
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: In emission tomography, iterative reconstruction is usually followed by a linear smoothing filter to make such images more appropriate for visual inspection and diagnosis by a physician. This will result in a global blurring of the images, smoothing across edges and possibly discarding valuable image information for detection tasks. The purpose of this study is to investigate which possible advantages a non-linear, edge-preserving postfilter could have on lesion detection in Ga-67 SPECT imaging. Image quality can be defined based on the task that has to be performed on the image. This study used LROC observer studies based on a dataset created by CPU-intensive Gate Monte Carlo simulations of a voxelized digital phantom. The filters considered in this study were a linear Gaussian filter, a bilateral filter, the Perona-Malik anisotropic diffusion filter and the Catté filtering scheme. The 3D MCAT software phantom was used to simulate the distribution of Ga-67 citrate in the abdomen. Tumor-present cases had a 1-cm diameter tumor randomly placed near the edges of the anatomical boundaries of the kidneys, bone, liver and spleen. Our data set was generated out of a single noisy background simulation using the bootstrap method, to significantly reduce the simulation time and to allow for a larger observer data set. Lesions were simulated separately and added to the background afterwards. These were then reconstructed with an iterative approach, using a sufficiently large number of MLEM iterations to establish convergence. The output of a numerical observer was used in a simplex optimization method to estimate an optimal set of parameters for each postfilter. No significant improvement was found for using edge-preserving filtering techniques over standard linear Gaussian filtering.
    Proc SPIE 03/2006;
  • [show abstract] [hide abstract]
    ABSTRACT: In this paper we introduce and evaluate a completely data-driven method to optimize both the linear expansion coefficients and the used temporal basis functions in dynamic PET reconstruction from list-mode data.We present the first results of our method using simulated 2D PET data. The time activity curves are modeled as a conic combination of B-splines. The B-splines are optimized with respect to their knot locations. We have combined the newly introduced method with our previously developed dynamic MLEM algorithm in order to further improve the likelihood. The results show that even one iteration of our newly developed algorithm can drastically increase the likelihood and the resulting images better represent the underlying TACs.
    Nuclear Instruments and Methods in Physics Research Section A Accelerators Spectrometers Detectors and Associated Equipment 01/2006; · 1.14 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: We validated the Monte Carlo package, GATE, for multi-pinhole gamma-ray imaging simulations using calculations and measurements. For these studies requiring large amount of photon histories, the newly designed cluster version of GATE was used. Accordingly that validated simulator allowed us then to evaluate design parameters for 123I-experiments on a prototype multi-pinhole SPECT camera for which analytical calculations were failing due to scatter of high-energy photons.
    Nuclear Instruments and Methods in Physics Research Section A Accelerators Spectrometers Detectors and Associated Equipment 01/2006; 569(2):359-363. · 1.14 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: In PET imaging, listmode is an alternative for projections to store the acquisition data with more accuracy. The reconstruction starting from the listmode format requires different approaches from projection based algorithms. The ML-EM algorithm was modified to start from the listmode format. For listmode data image degrading effects can be classified into two groups. Internal LOR effects (like attenuation, detection efficiency) do not lead to another LOR than the correct one, but only lead to a lower detection efficiency along the correct LOR. The correction for these effects can be done by using a sensitivity factor in the iterative algorithm. On the other hand there are also external LOR effects (scatter, randoms, resolution) which do lead to detection of wrong LORs. The correction for these effects can in principle be done by modifying the system matrix used in iterative reconstruction. For listmode reconstruction this system matrix can not be precalculated because of the very high number of possible LORs. Therefore we combined a simple model (Siddon) for the forward and backprojection with a depth dependent blurring function for the close range external LOR effects. It is shown how the resolution varies with the distance from the detector and how this effect can be modelled efficiently in combination with the Siddon algorithm. Effects like scatter and randoms are corrected by using an additive term for the forward projection. We show how the number of randoms along a LOR is estimated from the singles distribution and how this effect is corrected by the additive term in the listmode reconstruction.
    Biomedical Imaging, 2002. Proceedings. 2002 IEEE International Symposium on; 02/2002
  • [show abstract] [hide abstract]
    ABSTRACT: Gamma cameras used in PET mode are operating at high count rates, which leads to a high number of randoms in the acquired data. These data are often stored in listmode format to obtain more accurate reconstructions. We propose a randoms correction for an iterative list mode based reconstruction, which is based on the single photon distribution on each detector head. The method takes the spatial variation of the random coincidences into account. From the singles rates on each head the number of randoms in each voxel of the reconstruction is estimated. This is done by integrating the product of the singles over the solid angle formed by the voxel and both detectors. Camera rotation is taken Into account by integration this over all detector angles. This method only requires a quick singles scan prior or after the acquisition when there is unknown activity outside the FOV. When there is no activity outside the FOV or it is known the singles distribution on the detector head can be estimated from the reconstructed images. The randoms distribution can be subtracted after the reconstruction has been done
    Nuclear Science Symposium Conference Record, 2001 IEEE; 02/2001
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: Tomographic reconstruction from PET data is an ill-posed problem that requires regularization. Recently, Daubechies et al. [2] proposed an ℓ1 regularization of the wavelet coefficients that can be optimized using iterative thresh-olding schemes. In this paper, we extend this approach for the reconstruction of dynamic (spatio-temporal) PET data. Instead of using classical wavelets in the temporal dimension, we introduce exponential-spline wavelets that are specially tailored to model time activity curves (TACs) in PET. We show the usefulness of spatio-temporal regularization and the superior performance of E-spline wavelets over conventional Battle-Lemarié wavelets for a 1-D TAC fitting experiment and a tomographic reconstruction experiment.