J. Cabello

Comparison of basis functions for 3D PET reconstruction using a Monte Carlo system matrix

In emission tomography, iterative statistical methods are accepted as the reconstruction algorithms that achieve the best image quality. The accuracy of these methods relies partly on the quality of the system response matrix (SRM) that characterizes the scanner. The more physical phenomena included in the SRM, the higher the SRM quality, and therefore higher image quality is obtained from the reconstruction process. High-resolution small animal scanners contain as many as 10³-10⁴ small crystal pairs, while the field of view (FOV) is divided into hundreds of thousands of small voxels. These two characteristics have a significant impact on the number of elements to be calculated in the SRM. Monte Carlo (MC) methods have gained popularity as a way of calculating the SRM, due to the increased accuracy achievable, at the cost of introducing some statistical noise and long simulation times. In the work presented here the SRM is calculated using MC methods exploiting the cylindrical symmetries of the scanner, significantly reducing the simulation time necessary to calculate a high statistical quality SRM and the storage space necessary. The use of cylindrical symmetries makes polar voxels a convenient basis function. Alternatively, spherically symmetric basis functions result in improved noise properties compared to cubic and polar basis functions. The quality of reconstructed images using polar voxels, spherically symmetric basis functions on a polar grid, cubic voxels and post-reconstruction filtered polar and cubic voxels is compared from a noise and spatial resolution perspective. This study demonstrates that polar voxels perform as well as cubic voxels, reducing the simulation time necessary to calculate the SRM and the disk space necessary to store it. Results showed that spherically symmetric functions outperform polar and cubic basis functions in terms of noise properties, at the cost of slightly degraded spatial resolution, larger SRM file size and longer reconstruction times. However, we demonstrate that post-reconstruction smoothing, usually applied in emission imaging to reduce the level of noise, can produce a spatial resolution degradation of ~50%, while spherically symmetric basis functions produce a degradation of only ~6%, compared to polar and cubic voxels, at the same noise level. Therefore, the image quality trade-off obtained with blobs is higher than that obtained with cubic or polar voxels.

Development of a PET prototype with continuous LYSO crystals and monolithic SiPM matrices

A first prototype of a high resolution small animal PET scanner based on continuous LYSO scintillator crystals and silicon photomultiplier matrices has been developed at IFIC-Valencia, in collaboration with the University of Pisa and INFN Pisa. The prototype consists of two detector heads attached to a rotating stage. Each head is composed of a continuous 12 mm × 12 mm × 5 mm LYSO crystal painted white, coupled to a monolithic SiPM matrix. The matrices, developed at FBK-irst, are composed of 8×8 SiPM elements of 1.5 mm×1.4 mm size in a common substrate. The full characterization of the detector heads with different types of crystals has been carried out, and a method for determining the interaction position of the gamma-rays in the crystals has been successfully employed. Tomographic data have been acquired with the prototype, and images of different source distributions have been reconstructed with the Maximum Likelihood Expectation Maximization algorithm (MLEM). A FWHM close to 1 mm is obtained for one and two point-like sources.

Polar voxelization schemes combined with a Monte-Carlo based system matrix for image reconstruction in high resolution PET

Iterative methods are currently accepted as the gold standard image reconstruction methods in nuclear medicine. The quality of the final reconstructed image greatly depends on how good the physical processes are modelled in the System-Response-Matrix (SRM). Monte-Carlo based methods are a promising approach to calculate the SRM. However, the increasing granularity used in the detector and image space of high resolution small animal scanners has a direct impact on the time needed to calculate the matrix and its file size. The more physical processes are included in the SRM, or the better these processes are modelled, has a significant impact on the simulation time and the number of non-zero SRM elements. The use of polar voxels is an alternative to tackle this problem. In this work, a study on the performance and image quality of reconstructed images using polar voxels is compared to the traditional approach of cubic voxels, both based on a Monte-Carlo generated SRM. Several alternatives for polar voxelization, and comparison with cubic voxels are also studied in this work. The results obtained show that polar voxels produce reconstructed images with similar image quality, at the cost of reduced spatial resolution in the centre of the field of view (FOV), due to the elongated shape of the voxels in that region. This problem is of great importance, given that the centre of the FOV usually is the region of a PET scanner with highest sensitivity, and high spatial resolution in preclinical studies in this region is vital. A solution to this problem is proposed, introducing a different polar voxelization scheme for the central region of the FOV. It is demonstrated that the spatial resolution is fully recovered in this region, compared to Cartesian voxelization.

PET Reconstruction From Truncated Projections Using Total-Variation Regularization for Hadron Therapy Monitoring

Hadron therapy exploits the properties of ion beams to treat tumors by maximizing the dose released to the target and sparing healthy tissue. With hadron beams, the dose distribution shows a relatively low entrance dose which rises sharply at the end of the range, providing the characteristic Bragg peak that drops quickly thereafter. It is of critical importance in order not to damage surrounding healthy tissues and/or avoid targeting underdosage to know where the delivered dose profile ends-the location of the Bragg peak. During hadron therapy, short-lived β+-emitters are produced along the beam path, their distribution being correlated with the delivered dose. Following positron annihilation, two photons are emitted, which can be detected using a positron emission tomography (PET) scanner. The low yield of emitters, their short half-life, and the wash out from the target region make the use of PET, even only a few minutes after hadron irradiation, a challenging application. In-beam PET represents a potential candidate to estimate the distribution of β+-emitters during or immediately after irradiation, at the cost of truncation effects and degraded image quality due to the partial rings required of the PET scanner. Time-of-flight (ToF) information can potentially be used to compensate for truncation effects and to enhance image contrast. However, the highly demanding timing performance required in ToF-PET makes this option costly. Alternatively, the use of maximum-a-posteriori- expectation-maximization (MAP-EM), including total variation (TV) in the cost function, produces images with low noise, while preserving spatial resolution. In this paper, we compare data reconstructed with maximum-likelihood-expectation-maximization (ML-EM) and MAP-EM using TV as prior, and the impact of including ToF information, from data acquired with a complete and a partial-ring PET scanner, of simulated hadron beams interacting with a polymethyl methacrylate (PMMA) target. The results show that MAP-EM, in the absence of ToF information, produces lower noise images and more similar data compared to the simulated β+ distributions than ML-EM with ToF information in the order of 200-600 ps. The investigation is extended to the combination of MAP-EM and ToF information to study the limit of performance using both approaches.

High performance 3D PET reconstruction using spherical basis functions on a polar grid

Statistical iterative methods are a widely used method of image reconstruction in emission tomography. Traditionally, the image space is modelled as a combination of cubic voxels as a matter of simplicity. After reconstruction, images are routinely filtered to reduce statistical noise at the cost of spatial resolution degradation. An alternative to produce lower noise during reconstruction is to model the image space with spherical basis functions. These basis functions overlap in space producing a significantly large number of non-zero elements in the system response matrix (SRM) to store, which additionally leads to long reconstruction times. These two problems are partly overcome by exploiting spherical symmetries, although computation time is still slower compared to non-overlapping basis functions. In this work, we have implemented the reconstruction algorithm using Graphical Processing Unit (GPU) technology for speed and a precomputed Monte-Carlo-calculated SRM for accuracy. The reconstruction time achieved using spherical basis functions on a GPU was 4.3 times faster than the Central Processing Unit (CPU) and 2.5 times faster than a CPU-multi-core parallel implementation using eight cores. Overwriting hazards are minimized by combining a random line of response ordering and constrained atomic writing. Small differences in image quality were observed between implementations.

Position reconstruction in detectors based on continuous crystals coupled to silicon photomultiplier arrays

Sensitivity represents one of the major limitations for high resolution systems used in emission imaging for nuclear medicine. Currently, detectors are based on pixelated scintillators to detect ionizing radiation. There is a recent growing interest in continuous scintillators due to their increased sensitivity by eliminating insensitive areas in the detector crystals and reduced cost. To use such crystal an accurate position estimation algorithm is required to determine the location where photons interact, as opposed to pixelated scintillators, where the position is given by the crystal where the interaction took place. Additionally, including the depth where the photon interacted (DoI) inside the scintillator in the reconstruction algorithm, can be used to mitigate parallax effects. In this work we investigate the feasibility of using an existing analytical position estimation method applied to two different detectors designed to be used in two different imaging modalities: Positron Emission Tomography (PET) and Compton imaging. An LYSO crystal coupled to a 8×8 SiPM array, as one of the detector heads for the PET prototype, and a LaBr3 crystal coupled to a 4×4 SiPM array, as part of a Compton telescope comprised of several stacked detectors, is used for each application. Results show that submillimetric resolution is measured with both detectors in most of the studied positions, near the centre and close to the edges, in Monte Carlo simulations and experimentally. The measured FWHM for the PET detector is 0.6 mm while the FWHM measured for the detector used in the Compton camera is ∼0.7 mm. DoI measurements were taken only in simulations of the PET detector, where submillimetric bias was achieved and ∼1.6 mm FWHM was measured.

Simulated One Pass Listmode for fully 3D image reconstruction of Compton camera data

Image reconstruction for Compton camera data can be problematic due to the common trade-off between physically realistic models and speed of computation. In this investigation a novel method of system matrix calculation - Simulated One-Pass Listmode (SOPL) - is extended to incorporate Compton camera data. The method reduces the Cone Surface Response for the Compton camera to an ensemble of Siddon-rays and is conducted in two stages. As part of the ENVISION project for monitoring in hadron therapy, a continuous-crystal Lanthanum Bromide Compton camera has been developed and experimental data acquired. Continuous detection geometries are particularly susceptible to variation in both spatial and spectral resolution over the detection volume and so accurate yet flexible models of detection are particularly important. The SOPL-Compton method was applied via the Maximum Likelihood - Expectation Maximization algorithm to experimental data taken using the prototype device. In this investigation, detection modeling using SOPL-Compton in a two interaction Compton camera is validated and the incorporation of a shift-invariant image-space model confirmed as a useful modification to reduce computational expense. Finally experimental data taken using the prototype LaBr3 Compton camera provide confirmation of the SOPL-Compton approach to system modeling. Results indicate a fast, flexible and accurate algorithm that can easily be extended to alternate and novel detection geometries.

Application of Artificial Neural Network for Reducing Random Coincidences in PET

Positron Emission Tomography (PET) is based on the detection in coincidence of the two photons created in a positron annihilation. In conventional PET, this coincidence identification is usually carried out through a coincidence electronic unit. An accidental coincidence occurs when two photons arising from different annihilations are classified as a coincidence. Accidental coincidences are one of the main sources of image degradation in PET. Some novel systems allow coincidences to be selected post-acquisition in software, or in real time through a digital coincidence engine in an FPGA. These approaches provide the user with extra flexibility in the sorting process and allow the application of alternative coincidence sorting procedures. In this work a novel sorting procedure based on Artificial Neural Network (ANN) techniques has been developed. It has been compared to a conventional coincidence sorting algorithm based on a time coincidence window. The data have been obtained from Monte-Carlo simulations. A small animal PET scanner has been implemented to this end. The efficiency (the ratio of correct identifications) can be selected for both methods. In one case by changing the actual value of the coincidence window used, and in the other by changing a threshold at the output of the neural network. At matched efficiencies, the ANN-based method always produces a sorted output with a smaller random fraction. In addition, two differential trends are found: the conventional method presents a maximum achievable efficiency, while the ANN-based method is able to increase the efficiency up to unity, the ideal value, at the cost of increasing the random fraction. Images reconstructed using ANN sorted data (no compensation for randoms) present better contrast, and those image features which are more affected by randoms are enhanced. For the image quality phantom used in the paper, the ANN method decreases the spillover ratio by a factor of 18%.

Simulation study of Resistive-Plate-Chambers based PET for hadron-therapy monitoring

This is a preliminary study on the feasibility of using Resistive-Plate-Chambers (RPC) based Positron Emission Tomography (PET) for hadron-therapy monitoring. The imaging capabilities of the RPC gas detector are being investigated for PET. Their main advantages are excellent timing resolution, low cost and Depth Of Interaction information (DOI) due to their layered structure. Hadron-therapy (HT) aims at treating tumors by maximizing the dose released to the target and sparing healthy tissue. During irradiation with a hadron beam, fragments, positron emitting isotopes and gamma radiation are produced. This radiation coming from the tissue activation could be used for quality control of the treatment. Our work focuses on imaging the positron emitting isotopes using an RPC-based PET scanner. The low cost of RPC makes possible to enlarge the angular coverage of the scanner guaranteeing the higher sensitivity needed by the in-beam monitoring. In addition, the TOF capability and spatial resolution makes worth investigating it. A first protoype is currently under construction at CERN in the TERA group. For this study, Monte Carlo simulations by means of GATE were employed. Up on our knowledge, such gas detectors have not been simulated yet in GATE and the new version of GATE offers new tools for dosimetry. These are the main reasons of choosing GATE. A set of linear sources, simulating the delivery of a hadron beam, has proposed to evaluate the image quality for an ideal time resolution of the system and the TOF information is being considered in the reconstruction. For the full ring simulated system, no important differences were found between TOF and non-TOF images, only the contrast was always better for the TOF-images. However, further simulations should be performed adapting the system to the real situation (partial ring) for HT and lowering the detected events to a clinical number. Under these circumstances, we expect to be crucial the TOF capabilities of the RPC-PET.

First PET imaging results with continuous LYSO crystals and monolithic, 64-pixel SiPM matrices

The University of Pisa and INFN Pisa are developing a prototype of a small animal PET tomograph employing continuous LYSO crystals and silicon photom ultiplier (SiPM) matrices as photodetectors. The Center for Scientific and Technological Research (FBK-irst) has developed monolithic, 64 pixel SiPM matrices for their use in the PET scanner. The IFIC-Valencia collaborates in the characterization of the first detector heads, and in the construction of a first prototype for proof of concept. The detector heads are composed of continuous 12 mm × 12 mm × 5 mm LYSO crystals coupled to the matrices. The characterization of the detectors has been completed, and a first test prototype consisting of two rotating heads has been developed, assessing the feasibility of operating the system.