Pub Date : 2024-03-18DOI: 10.1109/TRPMS.2024.3378421
S. Di Giacomo;M. Ronchi;G. Borghi;D. R. Schaart;M. Carminati;C. Fiorini
Embedding signal processing in the front-end of radiation detectors represents an approach to cope with the growing complexity of nuclear imaging scanners with increasing field of view (i.e., higher number of channels). Machine learning (ML) offers a good compromise between intrinsic image reconstruction performance and computational power. While most hardware accelerators for ML are based on digital circuits and, thus, require the analog-to-digital conversion of all individual signals from photodetectors, an analog approach allows to streamline the pipeline. We present the study of an analog accelerator implementing a neural network (NN) with 42 neurons in a 0.35- ${mu }$ m CMOS process node. The specific target is the reconstruction of the position of interaction of gamma-rays in the scintillator crystal of Anger cameras used for PET and SPECT. This chip can be used stand-alone or monolithically integrated within the application specific integrated circuit (ASIC) for the filtering of current signals from arrays of silicon photomultipliers (SiPMs). Computation is performed in charge domain by means of crossbar arrays of programmable capacitor. The architecture of the 64-input ASIC and the training of the NN are presented, discussing the impact of weight quantization on 5 bits. From MATLAB and circuit simulations, consistent with ASIC topology and operations, the NN capabilities were tested using two different datasets, obtained from both simulated data and experimental data, both based on PET detector composed by a monolithic scintillator crystal readout by an $8times 8$ array of SiPMs. Simulations show an achievable spatial resolution better than 2-mm full-width-at-half-maximum with a 10-mm thick crystal, a max. count rate of 200kHz and the energy efficiency per inference is estimated to be of 93.5GOP/J, i.e., competitive with digital counterparts, with an energy consumption of 38nJ per inference and area of 23mm2.
随着视野范围的不断扩大（即通道数增加），核成像扫描仪的复杂性也在不断增加，将信号处理嵌入辐射探测器前端是应对这种情况的一种方法。机器学习（ML）是内在图像重建性能和计算能力之间的良好折衷。虽然大多数用于 ML 的硬件加速器都基于数字电路，因此需要对来自光电探测器的所有单独信号进行模数转换，但模拟方法可以简化流水线。我们介绍了对模拟加速器的研究，该加速器在 0.35- ${mu }$ m CMOS 工艺节点上实现了一个拥有 42 个神经元的神经网络 (NN)。具体目标是重建用于 PET 和 SPECT 的 Anger 相机闪烁晶体中伽马射线相互作用的位置。该芯片可独立使用，也可单片集成在专用集成电路（ASIC）中，用于过滤硅光电倍增管（SiPM）阵列发出的电流信号。计算是通过可编程电容器横条阵列在电荷域中进行的。介绍了 64 输入 ASIC 的结构和 NN 的训练，讨论了权重量化对 5 位的影响。根据 MATLAB 和电路仿真（与 ASIC 拓扑结构和操作一致），使用两个不同的数据集测试了 NN 的能力，这两个数据集均来自模拟数据和实验数据，均基于由单片闪烁晶体组成的 PET 检测器，该检测器由 8 美元/次的 SiPM 阵列读出。模拟结果表明，使用 10 毫米厚的晶体，可实现优于 2 毫米全宽半最大值的空间分辨率，最大计数率为 200kHz，每次推理的能效估计为 93.5GOP/J，即与数字同行相比具有竞争力，每次推理的能耗为 38nJ，面积为 23 平方毫米。
{"title":"Implementing an Integrated Neural Network for Real-Time Position Reconstruction in Emission Tomography With Monolithic Scintillators","authors":"S. Di Giacomo;M. Ronchi;G. Borghi;D. R. Schaart;M. Carminati;C. Fiorini","doi":"10.1109/TRPMS.2024.3378421","DOIUrl":"https://doi.org/10.1109/TRPMS.2024.3378421","url":null,"abstract":"Embedding signal processing in the front-end of radiation detectors represents an approach to cope with the growing complexity of nuclear imaging scanners with increasing field of view (i.e., higher number of channels). Machine learning (ML) offers a good compromise between intrinsic image reconstruction performance and computational power. While most hardware accelerators for ML are based on digital circuits and, thus, require the analog-to-digital conversion of all individual signals from photodetectors, an analog approach allows to streamline the pipeline. We present the study of an analog accelerator implementing a neural network (NN) with 42 neurons in a 0.35-\u0000<inline-formula> <tex-math>${mu }$ </tex-math></inline-formula>\u0000m CMOS process node. The specific target is the reconstruction of the position of interaction of gamma-rays in the scintillator crystal of Anger cameras used for PET and SPECT. This chip can be used stand-alone or monolithically integrated within the application specific integrated circuit (ASIC) for the filtering of current signals from arrays of silicon photomultipliers (SiPMs). Computation is performed in charge domain by means of crossbar arrays of programmable capacitor. The architecture of the 64-input ASIC and the training of the NN are presented, discussing the impact of weight quantization on 5 bits. From MATLAB and circuit simulations, consistent with ASIC topology and operations, the NN capabilities were tested using two different datasets, obtained from both simulated data and experimental data, both based on PET detector composed by a monolithic scintillator crystal readout by an \u0000<inline-formula> <tex-math>$8times 8$ </tex-math></inline-formula>\u0000 array of SiPMs. Simulations show an achievable spatial resolution better than 2-mm full-width-at-half-maximum with a 10-mm thick crystal, a max. count rate of 200kHz and the energy efficiency per inference is estimated to be of 93.5GOP/J, i.e., competitive with digital counterparts, with an energy consumption of 38nJ per inference and area of 23mm2.","PeriodicalId":46807,"journal":{"name":"IEEE Transactions on Radiation and Plasma Medical Sciences","volume":null,"pages":null},"PeriodicalIF":4.4,"publicationDate":"2024-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820220","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-17DOI: 10.1109/TRPMS.2024.3402272
Zixiang Chen;Yuxi Jin;Zhenxing Huang;Na Zhang;Kaiyi Liang;Guotao Quan;Dong Liang;Hairong Zheng;Zhanli Hu
Dynamic cerebral perfusion CT (PCT) is an effective imaging technique for the clinical diagnosis and therapy guidance of many kinds of cerebrovascular diseases (CVDs), but the large radiation dose imposed on a patient during repeated CT scans greatly limits its clinical applications. Achieving low-dose PCT imaging with the help of advanced and satisfactory imaging methods is needed. A kinetic-induced voxel-clustering filter (KVCF) is proposed in this work to help process noisy and distorted PCT images acquired from the low-dose CT scan protocols. In this new method, the intrinsic kinetic information of objective PCT images is extracted and effectively utilized to construct an image filter for every PCT frame. The new method is validated using both the simulated and clinical low-dose PCT data, and the peak signal-to-noise ratio (PSNR) and feature similarity (FSIM) are applied for quantitative evaluations of both the dynamic images and the calculated hemodynamic parametric maps. Compared to several existing methods, the proposed KVCF method produces the best qualitative and quantitative imaging effects. With satisfactory performance and high interpretability, KVCF is proven to be effective and implementable in the clinical low-dose PCT imaging tasks.
{"title":"Building a Kinetic Induced Voxel-Clustering Filter (KVCF) for Low-Dose Perfusion CT Imaging","authors":"Zixiang Chen;Yuxi Jin;Zhenxing Huang;Na Zhang;Kaiyi Liang;Guotao Quan;Dong Liang;Hairong Zheng;Zhanli Hu","doi":"10.1109/TRPMS.2024.3402272","DOIUrl":"https://doi.org/10.1109/TRPMS.2024.3402272","url":null,"abstract":"Dynamic cerebral perfusion CT (PCT) is an effective imaging technique for the clinical diagnosis and therapy guidance of many kinds of cerebrovascular diseases (CVDs), but the large radiation dose imposed on a patient during repeated CT scans greatly limits its clinical applications. Achieving low-dose PCT imaging with the help of advanced and satisfactory imaging methods is needed. A kinetic-induced voxel-clustering filter (KVCF) is proposed in this work to help process noisy and distorted PCT images acquired from the low-dose CT scan protocols. In this new method, the intrinsic kinetic information of objective PCT images is extracted and effectively utilized to construct an image filter for every PCT frame. The new method is validated using both the simulated and clinical low-dose PCT data, and the peak signal-to-noise ratio (PSNR) and feature similarity (FSIM) are applied for quantitative evaluations of both the dynamic images and the calculated hemodynamic parametric maps. Compared to several existing methods, the proposed KVCF method produces the best qualitative and quantitative imaging effects. With satisfactory performance and high interpretability, KVCF is proven to be effective and implementable in the clinical low-dose PCT imaging tasks.","PeriodicalId":46807,"journal":{"name":"IEEE Transactions on Radiation and Plasma Medical Sciences","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2024-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142143650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-15DOI: 10.1109/TRPMS.2024.3401391
Riccardo Latella;Antonio J. Gonzalez;José M. Benlloch;Paul Lecoq;Georgios Konstantinou
The signal-to-noise ratio in positron emission tomography (PET) improves with precise timing resolution. PET systems enabling the capability of time-of-flight (ToF) are nowadays available. This study assesses various data configurations, comparing the obtained timing performances applicable to time-of-flight positron emission tomography (ToF-PET) systems. Different readout configurations were evaluated together with silicon photomultipliers (SiPMs) photosensors from the Fondazione Bruno Kessler (FBK), with and without the so-called metal trench (MT) technology. The tests were carried out with scintillation crystals of $3times 3times $ 5 mm 3 (LYSO:Ce,Ca) from SIPAT. Two onboard FPGA-based systems, namely, the Felix time-to-digital converter (TDC) from Tediel S.r.l. and the ASIC-based FastIC from the University of Barcelona, along with custom-made high-frequency electronics (CM-HF), were compared. Considering only photopeak events, the best-coincidence timing resolution (CTR) results obtained were 71 ps with the MT SiPMs. This result worsened to 88 ps with the old version of the same device that does not include the MT technology (called HD). The results demonstrate substantial CTR improvements when MT SiPMs were used across the different scenarios, resulting in a timing improvement in the 10 to 45-ps range compared to HD SiPMs. Notably, the Felix TDC achieved sub-100-ps timing results, emphasizing the potential of FPGA technology in ToF-PET applications. Moreover, the fully passive version of the CM-HF connected to the MT SiPMs shows only a degradation of 8-ps difference compared to the version using amplifiers. The novel MT-type SiPMs promise superior timing performance, enhancing accuracy and efficiency in PET imaging systems.
{"title":"Comparative Analysis of Data Acquisition Setups for Fast Timing in ToF-PET Applications","authors":"Riccardo Latella;Antonio J. Gonzalez;José M. Benlloch;Paul Lecoq;Georgios Konstantinou","doi":"10.1109/TRPMS.2024.3401391","DOIUrl":"https://doi.org/10.1109/TRPMS.2024.3401391","url":null,"abstract":"The signal-to-noise ratio in positron emission tomography (PET) improves with precise timing resolution. PET systems enabling the capability of time-of-flight (ToF) are nowadays available. This study assesses various data configurations, comparing the obtained timing performances applicable to time-of-flight positron emission tomography (ToF-PET) systems. Different readout configurations were evaluated together with silicon photomultipliers (SiPMs) photosensors from the Fondazione Bruno Kessler (FBK), with and without the so-called metal trench (MT) technology. The tests were carried out with scintillation crystals of \u0000<inline-formula> <tex-math>$3times 3times $ </tex-math></inline-formula>\u00005 mm\u0000<sup>3</sup>\u0000 (LYSO:Ce,Ca) from SIPAT. Two onboard FPGA-based systems, namely, the Felix time-to-digital converter (TDC) from Tediel S.r.l. and the ASIC-based FastIC from the University of Barcelona, along with custom-made high-frequency electronics (CM-HF), were compared. Considering only photopeak events, the best-coincidence timing resolution (CTR) results obtained were 71 ps with the MT SiPMs. This result worsened to 88 ps with the old version of the same device that does not include the MT technology (called HD). The results demonstrate substantial CTR improvements when MT SiPMs were used across the different scenarios, resulting in a timing improvement in the 10 to 45-ps range compared to HD SiPMs. Notably, the Felix TDC achieved sub-100-ps timing results, emphasizing the potential of FPGA technology in ToF-PET applications. Moreover, the fully passive version of the CM-HF connected to the MT SiPMs shows only a degradation of 8-ps difference compared to the version using amplifiers. The novel MT-type SiPMs promise superior timing performance, enhancing accuracy and efficiency in PET imaging systems.","PeriodicalId":46807,"journal":{"name":"IEEE Transactions on Radiation and Plasma Medical Sciences","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2024-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10531074","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142143677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Molecular radiotherapy is a treatment modality that requires personalized dosimetry for efficient treatment and reduced toxicity. Current clinical imaging systems and miniaturized gamma-cameras lack the necessary features for this task. In this article, we present the design and optimization of a mobile gamma-camera with a $10times 10$ cm2 field of view tailored for quantitative imaging during $^{131}text{I}$ therapy of thyroid diseases. The camera uses a monolithic $10times 10times 1$ cm3 CeBr3 scintillator coupled to a $16times 16$ SiPMs array and commercial electronics. It exhibits high imaging performance with an intrinsic spatial resolution (SR) of 1.15-mm FWHM, an energy resolution of 8% FWHM at 356 keV and negligible deadtime up to 150 kcps. Images are reconstructed in real time using a convolutional neural network. The manufacturing method of tungsten collimators and shielding was optimized using laser 3-D printing to achieve an effective density of 97% that of bulk tungsten. Their geometry was adjusted with Monte-Carlo simulations in order to reduce septal penetration and scattering and optimize the signal-to-noise ratio at short times after treatment administration. Two high-energy parallel-hole collimators with high sensitivity or very high SR were designed for treatment planning and post-treatment control. The fully operational gamma-camera will soon be clinically assessed.
{"title":"A High-Resolution Portable Gamma-Camera for Estimation of Absorbed Dose in Molecular Radiotherapy","authors":"T. Bossis;M.-A. Verdier;C. Trigila;L. Pinot;F. Bouvet;A. Blot;H. Ramarijaona;T. Beaumont;D. Broggio;O. Caselles;S. Zerdoud;L. Ménard","doi":"10.1109/TRPMS.2024.3376826","DOIUrl":"https://doi.org/10.1109/TRPMS.2024.3376826","url":null,"abstract":"Molecular radiotherapy is a treatment modality that requires personalized dosimetry for efficient treatment and reduced toxicity. Current clinical imaging systems and miniaturized gamma-cameras lack the necessary features for this task. In this article, we present the design and optimization of a mobile gamma-camera with a \u0000<inline-formula> <tex-math>$10times 10$ </tex-math></inline-formula>\u0000 cm2 field of view tailored for quantitative imaging during \u0000<inline-formula> <tex-math>$^{131}text{I}$ </tex-math></inline-formula>\u0000 therapy of thyroid diseases. The camera uses a monolithic \u0000<inline-formula> <tex-math>$10times 10times 1$ </tex-math></inline-formula>\u0000 cm3 CeBr3 scintillator coupled to a \u0000<inline-formula> <tex-math>$16times 16$ </tex-math></inline-formula>\u0000 SiPMs array and commercial electronics. It exhibits high imaging performance with an intrinsic spatial resolution (SR) of 1.15-mm FWHM, an energy resolution of 8% FWHM at 356 keV and negligible deadtime up to 150 kcps. Images are reconstructed in real time using a convolutional neural network. The manufacturing method of tungsten collimators and shielding was optimized using laser 3-D printing to achieve an effective density of 97% that of bulk tungsten. Their geometry was adjusted with Monte-Carlo simulations in order to reduce septal penetration and scattering and optimize the signal-to-noise ratio at short times after treatment administration. Two high-energy parallel-hole collimators with high sensitivity or very high SR were designed for treatment planning and post-treatment control. The fully operational gamma-camera will soon be clinically assessed.","PeriodicalId":46807,"journal":{"name":"IEEE Transactions on Radiation and Plasma Medical Sciences","volume":null,"pages":null},"PeriodicalIF":4.4,"publicationDate":"2024-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10472319","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
X-ray spectrum plays an essential role in CT applications. Since it is difficult to measure X-ray spectrum directly in practice, X-ray spectrum is always indirectly obtained by using transmission measurements through a calibration phantom of known thickness and materials. These methods are independent of CT image reconstruction and bring extra cost. To solve this problem, we propose a parametric physical model-based X-ray spectrum estimation algorithm for CT imaging. First, an X-ray spectrum model with six parameters is proposed, which is derived from the X-ray imaging physics. Second, a template image containing different material components can be obtained by segmenting CT reconstructed images with a simple method. And the estimated projections can be calculated by reprojecting the template image with the proposed spectrum model. Finally, the six model parameters can be solved by iteratively minimizing the error between the estimated projection and real measurements. The effectiveness of the proposed method has been validated on both simulated and real data. Experimental results demonstrate that the proposed method can estimate the accurate spectra at different energies and provide a good reconstruction of characteristic radiations without extra cost.
X 射线光谱在 CT 应用中起着至关重要的作用。由于在实践中很难直接测量 X 射线频谱，因此总是通过已知厚度和材料的校准模型进行透射测量，从而间接获得 X 射线频谱。这些方法与 CT 图像重建无关，而且会带来额外成本。为解决这一问题，我们提出了一种基于参数物理模型的 CT 成像 X 射线光谱估算算法。首先，我们从 X 射线成像物理学出发，提出了具有六个参数的 X 射线光谱模型。其次，可以通过简单的方法分割 CT 重建图像，获得包含不同材料成分的模板图像。然后，利用所提出的光谱模型对模板图像进行再投影，即可计算出估计投影。最后，通过迭代最小化估计投影与实际测量之间的误差，即可求解六个模型参数。建议方法的有效性已在模拟数据和真实数据上得到验证。实验结果表明，所提出的方法可以估算出不同能量下的精确光谱，并在不增加额外成本的情况下提供良好的特征辐射重建。
{"title":"A Parametric Physical Model-Based X-Ray Spectrum Estimation Approach for CT Imaging","authors":"Shaojie Chang;Chaoyang Zhang;Xuanqin Mou;Qiong Xu;Lijun He;Xi Chen","doi":"10.1109/TRPMS.2024.3374702","DOIUrl":"https://doi.org/10.1109/TRPMS.2024.3374702","url":null,"abstract":"X-ray spectrum plays an essential role in CT applications. Since it is difficult to measure X-ray spectrum directly in practice, X-ray spectrum is always indirectly obtained by using transmission measurements through a calibration phantom of known thickness and materials. These methods are independent of CT image reconstruction and bring extra cost. To solve this problem, we propose a parametric physical model-based X-ray spectrum estimation algorithm for CT imaging. First, an X-ray spectrum model with six parameters is proposed, which is derived from the X-ray imaging physics. Second, a template image containing different material components can be obtained by segmenting CT reconstructed images with a simple method. And the estimated projections can be calculated by reprojecting the template image with the proposed spectrum model. Finally, the six model parameters can be solved by iteratively minimizing the error between the estimated projection and real measurements. The effectiveness of the proposed method has been validated on both simulated and real data. Experimental results demonstrate that the proposed method can estimate the accurate spectra at different energies and provide a good reconstruction of characteristic radiations without extra cost.","PeriodicalId":46807,"journal":{"name":"IEEE Transactions on Radiation and Plasma Medical Sciences","volume":null,"pages":null},"PeriodicalIF":4.4,"publicationDate":"2024-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820222","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-12DOI: 10.1109/TRPMS.2024.3368802
Georgios Konstantinou;Lei Zhang;Daniel Bonifacio;Riccardo Latella;Jose Maria Benlloch;Antonio J. Gonzalez;Paul Lecoq
In this study, we propose and examine a unique semimonolithic metascintillator (SMMS) detector design, where slow scintillators (BGO or LYSO) are split into thin slabs and read by an array of SiPM, offering depth-of-interaction (DOI) information. These are alternated with thin segmented fast scintillators (plastic EJ232 or EJ232Q), also read by single SiPMs, which provides pixel-level coincidence time resolution (CTR). The structure combines layers of slow scintillators of size $0.3times 25.5times $ (15 or 24) mm3 with fast scintillators of size $0.1times 3.1times $ (15 or 24) mm3. We use a Monte Carlo Gate simulation to gauge this novel semimonolithic detector’s performance. We found that the time resolution of SMMS is comparable to pixelated metascintillator designs with the same materials. For example, a 15-mm deep LYSO-based SMMS yielded a CTR of 121 ps before applying timewalk correction (after correction, 107-ps CTR). The equivalent BGO-based SMMS presented a CTR of 241 ps, which is a 15% divergence from metascintillator pixel experimental findings from previous works. We also applied neural networks to the photon distributions and timestamps recorded at the SiPM array, following guidelines on semimonolithic detectors. This led to determining the DOI with less than 3-mm precision and a confidence level of 0.85 in the best case, plus more than 2 standard deviations accuracy in reconstructing energy sharing and interaction energy. In summary, neural network prediction capabilities outperform standard energy calculation methods or any analytical approach on energy sharing, thanks to the improved understanding of photon distribution.
{"title":"Semi-Monolithic Meta-Scintillator Simulation Proof-of-Concept, Combining Accurate DOI and TOF","authors":"Georgios Konstantinou;Lei Zhang;Daniel Bonifacio;Riccardo Latella;Jose Maria Benlloch;Antonio J. Gonzalez;Paul Lecoq","doi":"10.1109/TRPMS.2024.3368802","DOIUrl":"https://doi.org/10.1109/TRPMS.2024.3368802","url":null,"abstract":"In this study, we propose and examine a unique semimonolithic metascintillator (SMMS) detector design, where slow scintillators (BGO or LYSO) are split into thin slabs and read by an array of SiPM, offering depth-of-interaction (DOI) information. These are alternated with thin segmented fast scintillators (plastic EJ232 or EJ232Q), also read by single SiPMs, which provides pixel-level coincidence time resolution (CTR). The structure combines layers of slow scintillators of size \u0000<inline-formula> <tex-math>$0.3times 25.5times $ </tex-math></inline-formula>\u0000 (15 or 24) mm3 with fast scintillators of size \u0000<inline-formula> <tex-math>$0.1times 3.1times $ </tex-math></inline-formula>\u0000(15 or 24) mm3. We use a Monte Carlo Gate simulation to gauge this novel semimonolithic detector’s performance. We found that the time resolution of SMMS is comparable to pixelated metascintillator designs with the same materials. For example, a 15-mm deep LYSO-based SMMS yielded a CTR of 121 ps before applying timewalk correction (after correction, 107-ps CTR). The equivalent BGO-based SMMS presented a CTR of 241 ps, which is a 15% divergence from metascintillator pixel experimental findings from previous works. We also applied neural networks to the photon distributions and timestamps recorded at the SiPM array, following guidelines on semimonolithic detectors. This led to determining the DOI with less than 3-mm precision and a confidence level of 0.85 in the best case, plus more than 2 standard deviations accuracy in reconstructing energy sharing and interaction energy. In summary, neural network prediction capabilities outperform standard energy calculation methods or any analytical approach on energy sharing, thanks to the improved understanding of photon distribution.","PeriodicalId":46807,"journal":{"name":"IEEE Transactions on Radiation and Plasma Medical Sciences","volume":null,"pages":null},"PeriodicalIF":4.4,"publicationDate":"2024-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820427","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-11DOI: 10.1109/TRPMS.2024.3372189
Aafke Christine Kraan;Alberto Del Guerra
In the last decades, important technological progress has been made to enhance the quality and efficiency of particle therapy treatments. Continuous improvements in dose delivery, treatment planning and verification techniques have led to higher-dose conformity and better sparing of healthy tissue. At the same time, particle therapy treatments are complex and much more expensive than conventional radiotherapy, and only highly specialized facilities can offer these treatments. Cost reduction is thus a strong drive behind technological developments in the field. The number of treatment facilities offering proton and carbon therapy has strongly grown in the last decades, and the amount of research efforts and innovations have increased continuously. From a technological perspective, advances in hardware are often accompanied by innovations in software and computation, and vice versa. In this review we will present a basic overview of technological advances in particle therapy hardware (accelerators, gantries, applications of superconductivity, treatment verification techniques), software (Monte Carlo simulations, treatment planning calculations), and studies toward clinical applications. By combining a broad selection of topics into a single review and by covering both proton and carbon therapy, we aim at providing the reader a unique overview of the evolution of various technologies developed for particle therapy.
{"title":"Technological Developments and Future Perspectives in Particle Therapy: A Topical Review","authors":"Aafke Christine Kraan;Alberto Del Guerra","doi":"10.1109/TRPMS.2024.3372189","DOIUrl":"https://doi.org/10.1109/TRPMS.2024.3372189","url":null,"abstract":"In the last decades, important technological progress has been made to enhance the quality and efficiency of particle therapy treatments. Continuous improvements in dose delivery, treatment planning and verification techniques have led to higher-dose conformity and better sparing of healthy tissue. At the same time, particle therapy treatments are complex and much more expensive than conventional radiotherapy, and only highly specialized facilities can offer these treatments. Cost reduction is thus a strong drive behind technological developments in the field. The number of treatment facilities offering proton and carbon therapy has strongly grown in the last decades, and the amount of research efforts and innovations have increased continuously. From a technological perspective, advances in hardware are often accompanied by innovations in software and computation, and vice versa. In this review we will present a basic overview of technological advances in particle therapy hardware (accelerators, gantries, applications of superconductivity, treatment verification techniques), software (Monte Carlo simulations, treatment planning calculations), and studies toward clinical applications. By combining a broad selection of topics into a single review and by covering both proton and carbon therapy, we aim at providing the reader a unique overview of the evolution of various technologies developed for particle therapy.","PeriodicalId":46807,"journal":{"name":"IEEE Transactions on Radiation and Plasma Medical Sciences","volume":null,"pages":null},"PeriodicalIF":4.4,"publicationDate":"2024-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10466736","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820217","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-08DOI: 10.1109/TRPMS.2024.3398360
Haiyan Wang;Han Jiang;Gefei Chen;Yu Du;Zhonglin Lu;Zhanli Hu;Greta S. P. Mok
Striatum segmentation on dopamine transporter (DaT) SPECT is necessary to quantify striatal uptake for Parkinson’s disease (PD), but is challenging due to the inferior resolution. This work proposes a cross-modality automatic striatum segmentation, estimating MR-derived striatal contours from clinical SPECT images using the deep learning (DL) methods. 123 I-Ioflupane DaT SPECT and T1-weighted MR images from 200 subjects with 152 PD and 48 healthy controls are analyzed from the Parkinson’s progression markers initiative database. SPECT and MR images are registered, and four striatal compartment contours are manually segmented from MR images as the label. DL methods including nnU-Net, U-Net, generative adversarial networks, and SPECT thresholding-based method are implemented for comparison. SPECT and MR label pairs are split into train, validation, and test groups (136:24:40). Dice, Hausdorff distance (HD) 95%, and relative volume difference (RVD), striatal binding ratio (SBR) and asymmetry index (ASI) are analyzed. Results show that nnU-Net achieves better Dice (~0.7), HD 95% (~1.8), and RVD (~0.1) as compared to other methods for all striatal compartments and whole striatum. For clinical PD evaluation, nnU-Net also yields strong SBR consistency (mean difference, −0.012) and ASI correlation (Pearson correlation coefficient, 0.81). The proposed DL-based cross-modality striatum segmentation method is feasible for clinical DaT SPECT in PD.
{"title":"Deep-Learning-Based Cross-Modality Striatum Segmentation for Dopamine Transporter SPECT in Parkinson’s Disease","authors":"Haiyan Wang;Han Jiang;Gefei Chen;Yu Du;Zhonglin Lu;Zhanli Hu;Greta S. P. Mok","doi":"10.1109/TRPMS.2024.3398360","DOIUrl":"https://doi.org/10.1109/TRPMS.2024.3398360","url":null,"abstract":"Striatum segmentation on dopamine transporter (DaT) SPECT is necessary to quantify striatal uptake for Parkinson’s disease (PD), but is challenging due to the inferior resolution. This work proposes a cross-modality automatic striatum segmentation, estimating MR-derived striatal contours from clinical SPECT images using the deep learning (DL) methods. \u0000<sup>123</sup>\u0000I-Ioflupane DaT SPECT and T1-weighted MR images from 200 subjects with 152 PD and 48 healthy controls are analyzed from the Parkinson’s progression markers initiative database. SPECT and MR images are registered, and four striatal compartment contours are manually segmented from MR images as the label. DL methods including nnU-Net, U-Net, generative adversarial networks, and SPECT thresholding-based method are implemented for comparison. SPECT and MR label pairs are split into train, validation, and test groups (136:24:40). Dice, Hausdorff distance (HD) 95%, and relative volume difference (RVD), striatal binding ratio (SBR) and asymmetry index (ASI) are analyzed. Results show that nnU-Net achieves better Dice (~0.7), HD 95% (~1.8), and RVD (~0.1) as compared to other methods for all striatal compartments and whole striatum. For clinical PD evaluation, nnU-Net also yields strong SBR consistency (mean difference, −0.012) and ASI correlation (Pearson correlation coefficient, 0.81). The proposed DL-based cross-modality striatum segmentation method is feasible for clinical DaT SPECT in PD.","PeriodicalId":46807,"journal":{"name":"IEEE Transactions on Radiation and Plasma Medical Sciences","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2024-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10525203","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142143654","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-08DOI: 10.1109/TRPMS.2024.3374207
Hao Sun;Yu Du;Ching-Ni Lin;Han Jiang;Wenbo Huang;Pai-Yi Chiu;Guang-Uei Hung;Lijun Lu;Greta S. P. Mok
This study aims to investigate robust attenuation correction (AC) by generating attenuation maps $(mu $ -maps) from nonattenuation-corrected (NAC) brain SPECT data using transfer learning (TL). Four sets of brain SPECT data ( $4times 30$ ) were retrospectively collected: S-TRODAT-1, S-ECD, G-TRODAT-1, and G-ECD. A 3-D attention-based conditional generative adversarial network was pretrained using 22 paired 3-D NAC SPECT images and corresponding CT $mu $ -maps for four patient groups. Various numbers ( $n,,=$ 4–22) of paired NAC SPECT and corresponding $mu $ -maps from S-TRODAT-1 were then used to fine-tune (FT) the other three pretrained deep learning (DL) networks, i.e., S-ECD, G-TRODAT-1, and G-ECD. All patients in S-TRODAT-1 group were tested on their own network (DL-AC), and on the pretrained models with FT (FT-AC) and without FT (NFT-AC). The FT-AC methods used 22 (FT22), 12 (FT12), 8 (FT8), and 4 (FT4) paired data for FT, respectively. Our results show that FT22 and FT12 could outperform DL-AC for cross-tracer S-ECD and cross-scanner G-TRODAT-1 using CT-based AC (CT-AC) as the reference. FT22 also outperforms DL-AC for cross-tracer+cross-scanner G-ECD. FT8 performs comparably to DL-AC, while FT4 is worse than DL-AC but still better than NAC and NFT-AC in each group. Attenuation map generation is feasible for brain SPECT based on cross-tracer and/or cross-scanner FT-AC using a smaller number of patient data. The FT-AC performance improves as the number of data used for FT increases.
{"title":"Cross-Tracer and Cross-Scanner Transfer Learning-Based Attenuation Correction for Brain SPECT","authors":"Hao Sun;Yu Du;Ching-Ni Lin;Han Jiang;Wenbo Huang;Pai-Yi Chiu;Guang-Uei Hung;Lijun Lu;Greta S. P. Mok","doi":"10.1109/TRPMS.2024.3374207","DOIUrl":"https://doi.org/10.1109/TRPMS.2024.3374207","url":null,"abstract":"This study aims to investigate robust attenuation correction (AC) by generating attenuation maps \u0000<inline-formula> <tex-math>$(mu $ </tex-math></inline-formula>\u0000-maps) from nonattenuation-corrected (NAC) brain SPECT data using transfer learning (TL). Four sets of brain SPECT data (\u0000<inline-formula> <tex-math>$4times 30$ </tex-math></inline-formula>\u0000) were retrospectively collected: S-TRODAT-1, S-ECD, G-TRODAT-1, and G-ECD. A 3-D attention-based conditional generative adversarial network was pretrained using 22 paired 3-D NAC SPECT images and corresponding CT \u0000<inline-formula> <tex-math>$mu $ </tex-math></inline-formula>\u0000-maps for four patient groups. Various numbers (\u0000<inline-formula> <tex-math>$n,,=$ </tex-math></inline-formula>\u0000 4–22) of paired NAC SPECT and corresponding \u0000<inline-formula> <tex-math>$mu $ </tex-math></inline-formula>\u0000-maps from S-TRODAT-1 were then used to fine-tune (FT) the other three pretrained deep learning (DL) networks, i.e., S-ECD, G-TRODAT-1, and G-ECD. All patients in S-TRODAT-1 group were tested on their own network (DL-AC), and on the pretrained models with FT (FT-AC) and without FT (NFT-AC). The FT-AC methods used 22 (FT22), 12 (FT12), 8 (FT8), and 4 (FT4) paired data for FT, respectively. Our results show that FT22 and FT12 could outperform DL-AC for cross-tracer S-ECD and cross-scanner G-TRODAT-1 using CT-based AC (CT-AC) as the reference. FT22 also outperforms DL-AC for cross-tracer+cross-scanner G-ECD. FT8 performs comparably to DL-AC, while FT4 is worse than DL-AC but still better than NAC and NFT-AC in each group. Attenuation map generation is feasible for brain SPECT based on cross-tracer and/or cross-scanner FT-AC using a smaller number of patient data. The FT-AC performance improves as the number of data used for FT increases.","PeriodicalId":46807,"journal":{"name":"IEEE Transactions on Radiation and Plasma Medical Sciences","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2024-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10461117","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141500371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-06DOI: 10.1109/TRPMS.2024.3397318
Mojtaba Jafaritadi;Emily Anaya;Garry Chinn;Jarrett Rosenberg;Tie Liang;Craig S. Levin
We present a context-aware generative deep learning framework to produce photon attenuation and scatter corrected (ASC) positron emission tomography (PET) images directly from nonattenuation and nonscatter corrected (NASC) images. We trained conditional generative adversarial networks (cGANs) on either single-modality (NASC) or multimodality (NASC+MRI) input data to map NASC images to pixel-wise continuously valued ASC PET images. We designed and evaluated four cGAN models including Pix2Pix, attention-guided cGAN (AG-Pix2Pix), vision transformer cGAN (ViT-GAN), and shifted window transformer cGAN (Swin-GAN). Retrospective 18F-fluorodeoxyglucose (18F-FDG) full-body PET images from 33 subjects were collected and analyzed. Notably, as a particular strength of this work, each patient in the study underwent both a PET/CT scan and a multisequence PET/MRI scan on the same day giving us a gold standard from the former as we investigate ASC for the latter. Quantitative analysis, evaluating image quality using peak signal-to-noise ratio (PSNR), multiscale structural similarity index (MS-SSIM), normalized mean-squared error (NRMSE), and mean absolute error (MAE) metrics, showed no significant impact of input type on PSNR ( $p=0.95$ ), MS-SSIM ( $p=0.083$ ), NRMSE ( $p=0.72$ ), or MAE ( $p=0.70$ ). For multimodal input data, Swin-GAN outperformed Pix2Pix ( $p=0.023$ ) and AG-Pix2Pix ( $p lt 0.001$ ), but not ViT-GAN ( $p=0.154$ ) in PSNR. Swin-GAN achieved significantly higher MS-SSIM than ViT-GAN ( $p=0.007$ ) and AG-Pix2Pix ( $p=0.002$ ). Multimodal Swin-GAN demonstrated reduced NRMSE and MAE compared to ViT-GAN ( $p=0.023$ and 0.031, respectively) and AG-Pix2Pix (both $p lt 0.001$ ), with marginal improvement over Pix2Pix ( $p lt 0.064$ ). The cGAN models, in particular Swin-GAN, consistently generated reliable and accurate ASC PET images, whether using multimodal or single-modal input data. The findings indicate that this methodology can be used to generate ASC data from standalone PET scanners or integrated PET/MRI systems, without relying on transmission scan-based attenuation maps.
{"title":"Context-Aware Transformer GAN for Direct Generation of Attenuation and Scatter Corrected PET Data","authors":"Mojtaba Jafaritadi;Emily Anaya;Garry Chinn;Jarrett Rosenberg;Tie Liang;Craig S. Levin","doi":"10.1109/TRPMS.2024.3397318","DOIUrl":"https://doi.org/10.1109/TRPMS.2024.3397318","url":null,"abstract":"We present a context-aware generative deep learning framework to produce photon attenuation and scatter corrected (ASC) positron emission tomography (PET) images directly from nonattenuation and nonscatter corrected (NASC) images. We trained conditional generative adversarial networks (cGANs) on either single-modality (NASC) or multimodality (NASC+MRI) input data to map NASC images to pixel-wise continuously valued ASC PET images. We designed and evaluated four cGAN models including Pix2Pix, attention-guided cGAN (AG-Pix2Pix), vision transformer cGAN (ViT-GAN), and shifted window transformer cGAN (Swin-GAN). Retrospective 18F-fluorodeoxyglucose (18F-FDG) full-body PET images from 33 subjects were collected and analyzed. Notably, as a particular strength of this work, each patient in the study underwent both a PET/CT scan and a multisequence PET/MRI scan on the same day giving us a gold standard from the former as we investigate ASC for the latter. Quantitative analysis, evaluating image quality using peak signal-to-noise ratio (PSNR), multiscale structural similarity index (MS-SSIM), normalized mean-squared error (NRMSE), and mean absolute error (MAE) metrics, showed no significant impact of input type on PSNR (\u0000<inline-formula> <tex-math>$p=0.95$ </tex-math></inline-formula>\u0000), MS-SSIM (\u0000<inline-formula> <tex-math>$p=0.083$ </tex-math></inline-formula>\u0000), NRMSE (\u0000<inline-formula> <tex-math>$p=0.72$ </tex-math></inline-formula>\u0000), or MAE (\u0000<inline-formula> <tex-math>$p=0.70$ </tex-math></inline-formula>\u0000). For multimodal input data, Swin-GAN outperformed Pix2Pix (\u0000<inline-formula> <tex-math>$p=0.023$ </tex-math></inline-formula>\u0000) and AG-Pix2Pix (\u0000<inline-formula> <tex-math>$p lt 0.001$ </tex-math></inline-formula>\u0000), but not ViT-GAN (\u0000<inline-formula> <tex-math>$p=0.154$ </tex-math></inline-formula>\u0000) in PSNR. Swin-GAN achieved significantly higher MS-SSIM than ViT-GAN (\u0000<inline-formula> <tex-math>$p=0.007$ </tex-math></inline-formula>\u0000) and AG-Pix2Pix (\u0000<inline-formula> <tex-math>$p=0.002$ </tex-math></inline-formula>\u0000). Multimodal Swin-GAN demonstrated reduced NRMSE and MAE compared to ViT-GAN (\u0000<inline-formula> <tex-math>$p=0.023$ </tex-math></inline-formula>\u0000 and 0.031, respectively) and AG-Pix2Pix (both \u0000<inline-formula> <tex-math>$p lt 0.001$ </tex-math></inline-formula>\u0000), with marginal improvement over Pix2Pix (\u0000<inline-formula> <tex-math>$p lt 0.064$ </tex-math></inline-formula>\u0000). The cGAN models, in particular Swin-GAN, consistently generated reliable and accurate ASC PET images, whether using multimodal or single-modal input data. The findings indicate that this methodology can be used to generate ASC data from standalone PET scanners or integrated PET/MRI systems, without relying on transmission scan-based attenuation maps.","PeriodicalId":46807,"journal":{"name":"IEEE Transactions on Radiation and Plasma Medical Sciences","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2024-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10521624","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141500327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：