Frank Kehren

Fully 3-D PET reconstruction with system matrix derived from point source measurements

The quality of images reconstructed by statistical iterative methods depends on an accurate model of the relationship between image space and projection space through the system matrix. The elements of the system matrix for the clinical Hi-Rez scanner were derived by processing the data measured for a point source at different positions in a portion of the field of view. These measured data included axial compression and azimuthal interleaving of adjacent projections. Measured data were corrected for crystal and geometrical efficiency. Then, a whole system matrix was derived by processing the responses in projection space. Such responses included both geometrical and detection physics components of the system matrix. The response was parameterized to correct for point source location and to smooth for projection noise. The model also accounts for axial compression (span) used on the scanner. The forward projector for iterative reconstruction was constructed using the estimated response parameters. This paper extends our previous work to fully three-dimensional. Experimental data were used to compare images reconstructed by the standard iterative reconstruction software and the one modeling the response function. The results showed that the modeling of the response function improves both spatial resolution and noise properties

First experimental results of time-of-flight reconstruction on an LSO PET scanner

Time-of-flight (TOF) positron emission tomography (PET) was studied and preliminarily developed in the 1980s, but the lack of a scintillator able to deliver at the same time proper time resolution and stopping power has prevented this technique from becoming widespread and commercially available. With the introduction of LSO in PET, TOF is now a feasible option. TOF reconstruction has been implemented in the CPS Hi-Rez PET scanner, both with 2D filtered-back-projection (FBP2D) and 3D ordered subset expectation maximization (OSEM3D). A new procedure has been introduced in the time alignment to compensate for the limited digital time resolution of the present electronics. A preliminary version of scatter correction for TOF has been devised and is presented. The measured time resolution of 1.2 ns (FWHM) allowed for a signal-to-noise ratio increase of about 50% in phantoms of about 40 cm transaxial size, or a gain larger than 2 in noise equivalent counts (NEC). TOF reconstruction has shown the expected improvement in SNR, both in simulation and experimental data. First experimental results show two improvements of TOF reconstruction over conventional (non-TOF) reconstruction: a lower noise level and a better capability to resolve structures deep inside large objects.

Completion of a Truncated Attenuation Image From the Attenuated PET Emission Data

Positron emission tomographs (PET) are currently almost exclusively designed as hybrid systems. The current standard is the PET/CT combination, while prototype PET/MRI systems are being studied by several research groups. One problem in these systems is that the transaxial field of view of the second system is smaller than that of the PET camera. The problem is limited for PET/CT, it is more pronounced in PET/MRI. Because this second system provides the image for attenuation correction, the smaller field of view causes truncation of the attenuation map. In this paper, we propose a maximum-a-posteriori algorithm for estimating the missing part of the attenuation map from the PET emission data.

PET reconstruction with system matrix derived from point source measurements

The quality of images reconstructed by statistical iterative methods depends on an accurate model of the relationship between image space and projection space through the system matrix. A method of acquiring the system matrix on the CPS Innovations the HiRez scanner was developed. The system matrix was derived by positioning the point source in the scanner field of view and processing the response in projection space. Such responses include geometrical and detection physics components of the system matrix. The response is parameterized to correct point source location and to smooth projection noise. Special attention was paid to span concepts of HiRez scanner. The projection operator for iterative reconstruction was constructed, taking into account estimated response parameters. The computer generated and acquired data were used to compare reconstruction obtained by the HiRez standard software and produced by better modeling. Results showed that the better resolution and noise property can be achieved.

SPMD cluster-based parallel 3D OSEM

This study empirically compares two approaches to parallel 3D OSEM that differ as to whether calculations are assigned to nodes by projection number or by transaxial plane number. For projection space decomposition (PSD), the forward projection is completely parallel, but backprojection requires a slow image synchronization. For image space decomposition (ISD), the communication associated with forward projection can be overlapped with calculation, and the communication associated with backprojection is more efficient. To compare these methods, an implementation of 3D OSEM for three PET scanners is developed that runs on an experimental, 9-node, 18-processor cluster computer. For selected benchmarks, both methods exhibit speedups in excess of 8 for 9 nodes, and comparable performance for the tested range of cluster sizes.

Continuous bed motion on clinical scanner: design, data correction, and reconstruction

With the addition of variable bed speed capabilities, continuous bed motion (CBM) becomes an acquisition generalization for axial sensitivity modeling and therefore a flexible tool in axial local image quality adjustment. In this paper, we describe the principles behind CBM mode planning and its influence on image reconstruction. The data correction method underwent the most changes compared to the commonly used step and shoot (S&S) mode. The CBM normalization array accommodates for activity decay, dead time correction, and the fact that various detector pairs acquire the same data for different durations. The normalization array is computed by simulating the movement of an object through the scanner, assisted by the monitoring of basic scanner acquisition parameters such as the singles rate. The sensitivity, which is the number of counts acquired per image plane, is an intrinsic part of normalization computing. Basic equations to estimate acquisition time at matched sensitivity between single speed CBM and S&S mode are presented. The CBM feature was implemented on a Siemens clinical scanner and initial studies of phantoms and patients are presented. The equivalence of single speed CBM and S&S image quality is demonstrated.

Incorporation of Axial System Response in Iterative Reconstruction from Axially Compressed Data of a Cylindrical Scanner using On-the-Fly Computing

The quality of images reconstructed by statistical iterative methods depends on an accurate model of the relationship between image space and projection space. Fast reconstruction is also an important aspect in the clinical environment. In previous work, we incorporated measured and modeled system response into iterative reconstruction. Since axial compression (span) was used in projection data, axial blurring was depth-dependent. We use pre-computed and stored axial components for a scanner without translation axial symmetry. For a scanner with translation symmetry, a fast line-of-response (LOR) line integral projector is available. In this case, axial blurring due to both geometrical and physics effects is implemented differently. Forward projection of axially blurred images is performed on-the-fly into LOR space, followed by axial compression of LOR data into scanner projection data formats. Our LOR projector is based on Josephs method. The translation symmetry of the scanner is essential for its efficient and fast computing performance. LOR projection and span assembling are performed independently for each azimuthal angle and result in small additional memory allocation. Results show that an acceleration of the computing of the axial blurring is achievable without any component storage but with similar image quality.

Continuous bed motion data processing for a resolution LSO PET/CT scanner

Continuous whole-body PET scanning, continuous bed motion (CBM) acquisition has a number of advantages over the traditional step-and-shoot (SS) mode. Strengths of CBM include a uniform axial signal-to-noise ratio, continuous sampling in the axial direction that reduces resolution artifacts, reduction of noise from detector normalization, and reduction of sensitivity to small patient movements. This work highlights the acquisition and data handling methodology that was implemented for a series of phantoms and over 40 patient studies acquired on a high resolution, 16-slice LSO combined PET/CT scanner (CPS Innovations, Knoxville, TN). CBM data were acquired in 32-bit listmode with the bed moving at a constant speed of typically 0.6 mm/s to match the acquisition time per plane of the SS mode. CBM data were processed using the novel, virtual scanner concept that can be applied to a scanner of any axial length. For the high resolution scanner, the LSO PET detectors are arranged in a truncated spherical geometry and therefore normalization and geometrical corrections are applied on an event-by-event basis during histogramming of the 32-bit listmode data. Scatter correction is calculated on the entire image volume, in contrast to the SS mode where scatter is estimated for each bed position. The final 3D data set was reconstructed using ordinary Poisson OSEM3D. This paper will present results from phantom studies and compare clinical patient scans acquired in both SS and CBM modes

Application of discrete data consistency conditions for selecting regularization parameters in PET attenuation map reconstruction.

Simultaneous emission and transmission measurement is appealing in PET due to the matching of geometrical conditions between emission and transmission and reduced acquisition time for the study. A potential problem remains: when transmission statistics are low, attenuation correction could be very noisy. Although noise in the attenuation map can be controlled through regularization during statistical reconstruction, the selection of regularization parameters is usually empirical. In this paper, we investigate the use of discrete data consistency conditions (DDCC) to optimally select one or two regularization parameters. The advantages of the method are that the reconstructed attenuation map is consistent with the emission data and that it accounts for particularity in the emission reconstruction algorithm and acquisition geometry. The methodology is validated using a computer-generated whole-body phantom for both emission and transmission, neglecting random events and scattered radiation. MAP-TR was used for attenuation map reconstruction, while 3D OS-EM is used for estimating the emission image. The estimation of regularization parameters depends on the resolution of the emission image controlled by the number of iterations in OS-EM. The computer simulation shows that, on one hand, DDCC regularized attenuation map reduces propagation of the transmission scan noise to the emission image, while on the other hand DDCC prevents excessive attenuation map smoothing that could result in resolution mismatch artefacts between emission and transmission.

Implementation of time-of-flight on CPS HiRez PET scanner

Time-of-flight (TOF) positron emission tomography (PET) was studied and preliminarily developed in the 80s, but the lack of a scintillator able to deliver at the same time proper time resolution and stopping power has prevented this technique from becoming widespread and commercially available. The introduction of LSO in PET has made TOF-PET a feasible option: TOF reconstruction is being implemented in CPS HiRez PET scanner, both in Filtered-Back-Projection and OSEM3D. The measured time resolution of 1.2 ns (FWHM) allows for a signal-to-noise ratio increase of about 50% in patients of about 40 cm transaxial size. TOF reconstruction has shown the expected improvement in SNR, both in simulation and experimental data. In iterative reconstruction we observed that TOF-PET has a much faster convergence compared to conventional reconstruction and better consistency improves the SNR ratio