Optical simulation study on the spatial resolution of a thick monolithic PET detector

Stockhoff, Mariele; Holen, Roel Van; Vandenberghe, Stefaan

doi:10.1088/1361-6560/ab3b83

Cited by 25 publications

(42 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other authors have reported CBCT scatter correction methods based on deep learning which operate in the image domain [27,60,84,140,152,155]. More specifically, they take a CBCT image as input and generate a synthetic CT image as output, i.e.…”

Section: Ai For Modelling Scattermentioning

confidence: 99%

Artificial Intelligence for Monte Carlo Simulation in Medical Physics

et al. 2021

View full text Add to dashboard Cite

Monte Carlo simulation of particle tracking in matter is the reference simulation method in the field of medical physics. It is heavily used in various applications such as 1) patient dose distribution estimation in different therapy modalities (radiotherapy, protontherapy or ion therapy) or for radio-protection investigations of ionizing radiation-based imaging systems (CT, nuclear imaging), 2) development of numerous imaging detectors, in X-ray imaging (conventional CT, dual-energy, multi-spectral, phase contrast … ), nuclear imaging (PET, SPECT, Compton Camera) or even advanced specific imaging methods such as proton/ion imaging, or prompt-gamma emission distribution estimation in hadrontherapy monitoring. Monte Carlo simulation is a key tool both in academic research labs as well as industrial research and development services. Because of the very nature of the Monte Carlo method, involving iterative and stochastic estimation of numerous probability density functions, the computation time is high. Despite the continuous and significant progress on computer hardware and the (relative) easiness of using code parallelisms, the computation time is still an issue for highly demanding and complex simulations. Hence, since decades, Variance Reduction Techniques have been proposed to accelerate the processes in a specific configuration. In this article, we review the recent use of Artificial Intelligence methods for Monte Carlo simulation in medical physics and their main associated challenges. In the first section, the main principles of some neural networks architectures such as Convolutional Neural Networks or Generative Adversarial Network are briefly described together with a literature review of their applications in the domain of medical physics Monte Carlo simulations. In particular, we will focus on dose estimation with convolutional neural networks, dose denoising from low statistics Monte Carlo simulations, detector modelling and event selection with neural networks, generative networks for source and phase space modelling. The expected interests of those approaches are discussed. In the second section, we focus on the current challenges that still arise in this promising field.

show abstract

Section: Ai For Modelling Scattermentioning

confidence: 99%

Artificial Intelligence for Monte Carlo Simulation in Medical Physics

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Indeed, detector-dependent factors, such as crystal size and crystal penetration during detection, and inherent limitations such as the non-collinearity of the annihilation photons are also present and can explain why a reduced positron range with an increasing magnetic field does not translate directly into improved image quality ( Herzog et al, 2010 ; Bertolli et al, 2016 ; Caribé et al, 2019 ; Wadhwa et al, 2020 ). However, in a preclinical setting with small diameter detector rings and crystal sizes, the impact of a reduced positron range on the PET image quality is expected to be much higher while the use of monolithic crystals or recordings of the depth of interaction in clinical PET systems can further enhance the PET resolution such that it becomes more sensitive to positron range effects ( Hammer et al, 1994 ; Stockhoff et al, 2019 ).…”

Section: Discussionmentioning

confidence: 99%

Monte Carlo Simulations of the GE Signa PET/MR for Different Radioisotopes

et al. 2020

Self Cite

View full text Add to dashboard Cite

NEMA characterization of PET systems is generally based on 18 F because it is the most relevant radioisotope for the clinical use of PET. 18 F has a half-life of 109.7 min and decays into stable 18 O via β+ emission with a probability of over 96% and a maximum positron energy of 0.633 MeV. Other commercially available PET radioisotopes, such as 82 Rb and 68 Ga have more complex decay schemes with a variety of prompt gammas, which can directly fall into the energy window and induce false coincidence detections by the PET scanner. Methods: Aim of this work was threefold: (A) Develop a GATE model of the GE Signa PET/MR to perform realistic and relevant Monte Carlo simulations (B) Validate this model with published sensitivity and Noise Equivalent Count Rate (NECR) data for 18 F and 68 Ga (C) Use the validated GATE-model to predict the system performance for other PET isotopes including 11 C, 15 O, 13 N, 82 Rb, and 68 Ga and to evaluate the effect of a 3T magnetic field on the positron range. Results: Simulated sensitivity and NECR tests performed with the GATE-model for different radioisotopes were in line with literature values. Simulated sensitivities for 18 F and 68 Ga were 21.2 and 19.0 /kBq, respectively, for the center position and 21.1 and 19.0 cps/kBq, respectively, for the 10 cm off-center position compared to the corresponding measured values of 21.8 and 20.0 cps/kBq for the center position and 21.1 and 19.6 cps/kBq for the 10 cm off-center position. In terms of NECR, the simulated peak NECR was 216.8 kcps at 17.40 kBq/ml for 18 F and 207.1 kcps at 20.10 kBq/ml for 68 Ga compared to the measured peak NECR of 216.8 kcps at 18.60 kBq/ml and 205.6 kcps at 20.40 kBq/ml for 18 F and 68 Ga, respectively. For 11 C, 13 N, and 15 O, results confirmed a peak NECR similar to 18 F with the effective activity concentration scaled by the inverse of the positron fraction. For 82 Rb, and 68 Ga, the peak NECR was lower than for 18 F while the corresponding activity concentrations were higher. For the higher energy positron emitters, the positron range was confirmed to be

show abstract

“…To evaluate the DOI estimation of both algorithms the predicted relative number of events in each of those layers is compared. We compare to (i) the theoretical number of events that we expect by the attenuation of the crystal according to Beer-Lambert Law and (ii) the results we previously obtained from optical simulations (Stockhoff et al 2019) modelling the same detector geometry, calibration procedure and MNN positioning. Note that in the simulations the DOI was evaluated for only the centre 10 x 10 mm 2 .…”

Section: Doi Estimationmentioning

confidence: 99%

High-resolution monolithic LYSO detector with 6-layer depth-of-interaction for clinical PET

et al. 2021

Self Cite

View full text Add to dashboard Cite

The system spatial resolution of whole-body positron emission tomography (PET) is limited to around 2 mm due to positron physics and the large diameter of the bore. To stay below this 'physics'limit a scintillation detector with an intrinsic spatial resolution of around 1.3 mm is needed. Currently used detector technology consists of arrays of 2.6-5 mm segmented scintillator pixels which are the dominant factor contributing to the system resolution. Pixelated detectors using smaller pixels exist but face major drawbacks in sensitivity, timing, energy resolution and cost. Monolithic continuous detectors, where the spatial resolution is determined by the shape of the light distribution on the photo detector array, are a promising alternative. Without having the drawbacks of pixelated detectors, monolithic ones can also provide depth-of-interaction (DOI) information. In this work we present a monolithic detector design aiming to serve high-resolution clinical PET systems while maintaining high sensitivity. A 50 x 50 x 16 mm 3 Lutetium-Yttrium oxyorthosilicate (LYSO) scintillation crystal with silicon photomultiplier (SiPM) back side readout is calibrated in singles mode by a collimated beam obtaining a reference dataset for the event positioning. A mean nearest neighbour (MNN) algorithm and an artificial neural network for positioning are compared. The targeted intrinsic detector resolution of 1.3 mm needed to reach a 2 mm resolution on system level was accomplished with both algorithms. The neural network achieved a mean spatial resolution of 1.14 mm FWHM for the whole detector and 1.02 mm in the centre (30 x 30 mm 2 ). The MNN algorithm performed slightly worse with 1.17 mm for the whole detector and 1.13 mm in the centre. The intrinsic DOI information will also result in uniform system spatial resolution over the full field of view.

show abstract

Optical simulation study on the spatial resolution of a thick monolithic PET detector

Cited by 25 publications

References 36 publications

Artificial Intelligence for Monte Carlo Simulation in Medical Physics

Artificial Intelligence for Monte Carlo Simulation in Medical Physics

Monte Carlo Simulations of the GE Signa PET/MR for Different Radioisotopes

High-resolution monolithic LYSO detector with 6-layer depth-of-interaction for clinical PET

Contact Info

Product

Resources

About