Search

Objective. This study addresses a fundamental limitation of in-beam positron emission tomography (IB-PET) in proton therapy: the lack of direct anatomical representation in the images it produces. We aim to overcome this shortcoming by pioneering the application of deep learning techniques to create synthetic control CT images (sCT) from combining IB-PET and planning CT scan data. Approach. We conducted simulations involving six patients who underwent irradiation with proton beams. Leveraging the architecture of a visual transformer (ViT) neural network, we developed a model to generate sCT images of these patients using the planning CT scans and the inter-fractional simulated PET activity maps during irradiation. To evaluate the model's performance, a comparison was conducted between the sCT images produced by the ViT model and the authentic control CT images-serving as the benchmark. Main results. The structural similarity index was computed at a mean value across all patients of 0.91, while the mean absolute error measured 22 Hounsfield Units (HU). Root mean squared error and peak signal-to-noise ratio values were 56 HU and 30 dB, respectively. The Dice similarity coefficient exhibited a value of 0.98. These values are comparable to or exceed those found in the literature. More than 70% of the synthetic morphological changes were found to be geometrically compatible with the ones reported in the real control CT scan. Significance. Our study presents an innovative approach to surface the hidden anatomical information of IB-PET in proton therapy. Our ViT-based model successfully generates sCT images from inter-fractional PET data and planning CT scans. Our model's performance stands on par with existing models relying on input from cone beam CT or magnetic resonance imaging, which contain more anatomical information than activity maps., (Creative Commons Attribution license.)

Objective . Monolithic scintillator crystals coupled to silicon photomultiplier (SiPM) arrays are promising detectors for PET applications, offering spatial resolution around 1 mm and depth-of-interaction information. However, their timing resolution has always been inferior to that of pixellated crystals, while the best results on spatial resolution have been obtained with algorithms that cannot operate in real-time in a PET detector. In this study, we explore the capabilities of monolithic crystals with respect to spatial and timing resolution, presenting new algorithms that overcome the mentioned problems. Approach. Our algorithms were tested first using a simulation framework, then on experimentally acquired data. We tested an event timestamping algorithm based on neural networks which was then integrated into a second neural network for simultaneous estimation of the event position and timestamp. Both algorithms are implemented in a low-cost field-programmable gate array that can be integrated in the detector and can process more than 1 million events per second in real-time. Results. Testing the neural network for the simultaneous estimation of the event position and timestamp on experimental data we obtain 0.78 2D FWHM on the ( x , y ) plane, 1.2 depth-of-interaction FWHM and 156 coincidence time resolution on a25mm×25mm×8mm×LYSO monolith read-out by 643mm×3mmHamamatsu SiPMs. Significance. Our results show that monolithic crystals combined with artificial intelligence can rival pixellated crystals performance for time-of-flight PET applications, while having better spatial resolution and DOI resolution. Thanks to the use of very light neural networks, event characterization can be done on-line directly in the detector, solving the issues of scalability and computational complexity that up to now were preventing the use of monolithic crystals in clinical PET scanners., (© 2022 Institute of Physics and Engineering in Medicine.)

Positron emission tomography is one of the most mature techniques for monitoring the particles range in hadron therapy, aiming to reduce treatment uncertainties and therefore the extent of safety margins in the treatment plan. In-beam PET monitoring has been already performed using inter-spill and post-irradiation data, i.e. while the particle beam is off or paused. The full beam acquisition procedure is commonly discarded because the particle spills abruptly increase the random coincidence rates and therefore the image noise. This is because random coincidences cannot be separated by annihilation photons originating from radioactive decays and cannot be corrected with standard random coincidence techniques due to the time correlation of the beam-induced background with the ion beam microstructure. The aim of this paper is to provide a new method to recover in-spill data to improve the images obtained with full-beam PET acquisitions. This is done by estimating the temporal microstructure of the beam and thus selecting input PET events that are less likely to be random ones. The PET detector we used was the one developed within the INSIDE project and tested at the CNAO synchrotron-based facility. The data were taken on a PMMA phantom irradiated with 72 MeV proton pencil beams. The obtained results confirm the possibility of improving the acquired PET data without any external signal coming from the synchrotron or ad hoc detectors.

Positron emission tomography (PET) iterative 3D reconstruction is a very computational demanding task. One of the main issues of the iterative reconstruction concerns the management of the system response matrix (SRM). The SRM models the relationship between the projection and the voxel space and its memory footprint can easily exceed hundreds of GB. Moreover, in order to make the reconstruction fast enough not to hinder its practical application, the SRM must be stored in the random access memory of the workstation used for the reconstruction. This issue is normally solved by implementing efficient storage schemes and by reducing the number of redundant patterns in the SRM through symmetries. However, finding a sufficient number of symmetries is often non-trivial and is typically performed using dedicated solutions that cannot be exported to different detectors and geometries. In this paper, an automatic approach to reduce the memory footprint of a pre-computed SRM is described. The proposed approach was named symmetry search algorithm (SSA) and consists in an algorithm that searches for some of the redundant patterns present in the SRM, leading to its lossy compression. This approach was built to detect translations, reflections and coordinates swap in voxel space. Therefore, it is particularly well suited for those scanners where some of the rotational symmetries are broken, e.g. small animal scanner where the modules are arranged in a polygonal ring made of few elements, and dual head planar PET systems. In order to validate this approach, the SSA is applied to the SRM of a preclinical scanner (the IRIS PET/CT). The data acquired by the scanner were reconstructed with a dedicated maximum likelihood estimation maximization algorithm with both the uncompressed and the compressed SRMs. The results achieved show that the information lost due to the SSA compression is negligible. Compression factors up to 52 when using the SSA together with manually inserted symmetries and up to 204 when using the SSA alone, can be obtained for the IRIS SRM. These results come without significant differences in the values and in the main quality metrics of the reconstructed images, i.e. spatial resolution and noise. Although the compression factors depend on the system considered, the SSA is applicable to any SRM and therefore it can be considered a general tool to reduce the footprint of a pre-computed SRM.

In vivo range monitoring techniques are necessary in order to fully take advantage of the high dose gradients deliverable in hadrontherapy treatments. Positron emission tomography (PET) scanners can be used to monitor beam-induced activation in tissues and hence measure the range. The INSIDE (Innovative Solutions for In-beam DosimEtry in Hadrontherapy) in-beam PET scanner, installed at the Italian National Center of Oncological Hadrontherapy (CNAO, Pavia, Italy) synchrotron facility, has already been successfully tested in vivo during a proton therapy treatment. We discuss here the system performance evaluation with carbon ion beams, in view of future in vivo tests. The work is focused on the analysis of activity images obtained with therapeutic treatments delivered to polymethyl methacrylate (PMMA) phantoms, as well as on the test of an innovative and robust Monte Carlo simulation technique for the production of reliable prior activity maps. Images are reconstructed using different integration intervals, so as to monitor the activity evolution during and after the treatment. Three procedures to compare activity images are presented, namely Pearson correlation coefficient, Beam's eye view and overall view. Images of repeated irradiations of the same treatments are compared to assess the integration time necessary to provide reproducible images. The range agreement between simulated and experimental images is also evaluated, so as to validate the simulation capability to provide sound prior information. The results indicate that at treatment end, or at most 20 s afterwards, the range measurement is reliable within 1-2 mm, when comparing both different experimental sessions and data with simulations. In conclusion, this work shows that the INSIDE in-beam PET scanner performance is promising towards its in vivo test with carbon ions.

During particle therapy irradiation, positron emitters with half-lives ranging from 2 to 20 min are generated from nuclear processes. The half-lives are such that it is possible either to detect the positron signal in the treatment room using an in-beam positron emission tomography (PET) system, right after the irradiation, or to quickly transfer the patient to a close PET/CT scanner. Since the activity distribution is spatially correlated with the dose, it is possible to use PET imaging as an indirect method to assure the quality of the dose delivery. In this work, we present a new dedicated PET system able to operate in-beam. The PET apparatus consists in two 10 cm × 10 cm detector heads. Each detector is composed of four scintillating matrices of 23 × 23 LYSO crystals. The crystal size is 1.9 mm × 1.9 mm × 16 mm. Each scintillation matrix is read out independently with a modularized acquisition system. The distance between the two opposing detector heads was set to 20 cm. The system has very low dead time per detector area and a 3 ns coincidence window, which is capable to sustain high single count rates and to keep the random counts relatively low. This allows a new full-beam monitoring modality that includes data acquisition also while the beam is on. The PET system was tested during the irradiation at the CATANA (INFN, Catania, Italy) cyclotron-based proton therapy facility. Four acquisitions with different doses and dose rates were analysed. In all cases the random to total coincidences ratio was equal or less than 25%. For each measurement we estimated the accuracy and precision of the activity range on a set of voxel lines within an irradiated PMMA phantom. Results show that the inclusion of data acquired during the irradiation, referred to as beam-on data, improves both the precision and accuracy of the range measurement with respect to data acquired only after irradiation. Beam-on data alone are enough to give precisions better than 1 mm when at least 5 Gy are delivered.