Pub Date : 2020-10-31DOI: 10.1109/NSS/MIC42677.2020.9508077
Andy Labella, Xinjie Cao, Wei Zhao, A. Goldan
Spatial performance of single-ended readout depth-encoding PET modules with multicrystal scintillator arrays that are n-to-1 coupled to readout pixels relies on the ability to identify the crystal where each gamma ray is absorbed based on light sharing patterns. Energy weighted average method is the most popular method for performing crystal identification in such detector modules. However, quantitative metrics that characterize flood histogram quality haven't yet been developed for this practical, cost-effective detector module configuration. In this work, we introduce a flood histogram quality metric that determines how well-separated the crystal identification clusters are when coupling multiple crystals to the same readout pixel. We compare the flood histogram quality between 4-to-1 coupled modules with a standard uniform glass light guide and our newly developed prismatoid light guide array, which is used in Prism-PET detector module configurations. Both modules consisted of 16 ×16 arrays of 1.4 × 1.4 × 20 mm3LYSO crystals coupled 4-to-1 to 3.2 × 3.2 mm2SiPM pixels. The Prism-PET module exhibited 40% better flood histogram quality than the uniform light guide module. Crystal clusters acquired at 5 different depths in both modules demonstrated how Prism-PET increases the depth-dependence of crystal contours, thus enhancing crystal separation. Our flood histogram quality metric is a quantitative measure that helps characterize high resolution single-ended readout modules with n-to-1 crystal-to-readout coupling.
{"title":"Flood Histogram Quality Metric for Light Sharing Depth-Encoding PET Modules","authors":"Andy Labella, Xinjie Cao, Wei Zhao, A. Goldan","doi":"10.1109/NSS/MIC42677.2020.9508077","DOIUrl":"https://doi.org/10.1109/NSS/MIC42677.2020.9508077","url":null,"abstract":"Spatial performance of single-ended readout depth-encoding PET modules with multicrystal scintillator arrays that are n-to-1 coupled to readout pixels relies on the ability to identify the crystal where each gamma ray is absorbed based on light sharing patterns. Energy weighted average method is the most popular method for performing crystal identification in such detector modules. However, quantitative metrics that characterize flood histogram quality haven't yet been developed for this practical, cost-effective detector module configuration. In this work, we introduce a flood histogram quality metric that determines how well-separated the crystal identification clusters are when coupling multiple crystals to the same readout pixel. We compare the flood histogram quality between 4-to-1 coupled modules with a standard uniform glass light guide and our newly developed prismatoid light guide array, which is used in Prism-PET detector module configurations. Both modules consisted of 16 ×16 arrays of 1.4 × 1.4 × 20 mm3LYSO crystals coupled 4-to-1 to 3.2 × 3.2 mm2SiPM pixels. The Prism-PET module exhibited 40% better flood histogram quality than the uniform light guide module. Crystal clusters acquired at 5 different depths in both modules demonstrated how Prism-PET increases the depth-dependence of crystal contours, thus enhancing crystal separation. Our flood histogram quality metric is a quantitative measure that helps characterize high resolution single-ended readout modules with n-to-1 crystal-to-readout coupling.","PeriodicalId":6760,"journal":{"name":"2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC)","volume":"97 1","pages":"1-3"},"PeriodicalIF":0.0,"publicationDate":"2020-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80754106","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 : 2020-10-31DOI: 10.1109/NSS/MIC42677.2020.9507755
J. Dickmann, F. Kamp, R. Schulte, K. Parodi, G. Dedes, G. Landry
We present an optimization algorithm for fluence-modulated proton computed tomography that allows prescribing spatially inhomogeneous dose and image noise distributions. This is particularly meaningful if proton CT images are used for particle therapy treatment planning and online adaptation, where only the region-or-interest (ROI) around the treatment beam path (i.e. the ROI) is relevant and imaging dose can be reduced elsewhere. This may allow for daily imaging at the treatment site with imaging doses that would not compromise the low dose to healthy tissue made possible by particle therapy. We investigate a typical treatment scenario with two beams and optimize dynamic fluence maps resulting in a dose reduction of 30% outside of the ROI. Increasing the magnitude of dose reduction inside a small volume around an organ-at-risk (OAR) brings the OAR dose to 62% below a scan without fluence modulation. This flexible optimization method may facilitate low-dose image guidance with proton CT.
{"title":"Joint Dose Minimization and Variance Optimization for Fluence-Modulated Proton CT","authors":"J. Dickmann, F. Kamp, R. Schulte, K. Parodi, G. Dedes, G. Landry","doi":"10.1109/NSS/MIC42677.2020.9507755","DOIUrl":"https://doi.org/10.1109/NSS/MIC42677.2020.9507755","url":null,"abstract":"We present an optimization algorithm for fluence-modulated proton computed tomography that allows prescribing spatially inhomogeneous dose and image noise distributions. This is particularly meaningful if proton CT images are used for particle therapy treatment planning and online adaptation, where only the region-or-interest (ROI) around the treatment beam path (i.e. the ROI) is relevant and imaging dose can be reduced elsewhere. This may allow for daily imaging at the treatment site with imaging doses that would not compromise the low dose to healthy tissue made possible by particle therapy. We investigate a typical treatment scenario with two beams and optimize dynamic fluence maps resulting in a dose reduction of 30% outside of the ROI. Increasing the magnitude of dose reduction inside a small volume around an organ-at-risk (OAR) brings the OAR dose to 62% below a scan without fluence modulation. This flexible optimization method may facilitate low-dose image guidance with proton CT.","PeriodicalId":6760,"journal":{"name":"2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC)","volume":"117 4 1","pages":"1-3"},"PeriodicalIF":0.0,"publicationDate":"2020-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75856098","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 : 2020-10-31DOI: 10.1109/NSS/MIC42677.2020.9507871
S. Bheesette, M. Turqueti
Accurate magnetic field measurements are fundamental to the construction, testing, and certification of magnetic systems. Often, in high accuracy systems, the measurement technique and its implementation may involve a considerable effort. One such example of this type of system is undulators for light sources. Advanced undulators require several magnetic measurements at different stages during their construction. Every magnet block, composed of several magnetic poles, must be measured individually and sorted based on the magnetic moment results. There are two degrees of freedom for each pole. First, for tuning the vertical field, a pole may be moved, and, second, the local gap formed by a top and bottom pole may also be adjusted for vertical and horizontal field errors. Usually, undulators are assembled with a collection of periodic blocks surveyed to assess their accurate positions. The final process of fine-tuning the undulators requires the magnetic measurements of the whole assembly.
{"title":"Electron Gun-Based Magnetic Probe","authors":"S. Bheesette, M. Turqueti","doi":"10.1109/NSS/MIC42677.2020.9507871","DOIUrl":"https://doi.org/10.1109/NSS/MIC42677.2020.9507871","url":null,"abstract":"Accurate magnetic field measurements are fundamental to the construction, testing, and certification of magnetic systems. Often, in high accuracy systems, the measurement technique and its implementation may involve a considerable effort. One such example of this type of system is undulators for light sources. Advanced undulators require several magnetic measurements at different stages during their construction. Every magnet block, composed of several magnetic poles, must be measured individually and sorted based on the magnetic moment results. There are two degrees of freedom for each pole. First, for tuning the vertical field, a pole may be moved, and, second, the local gap formed by a top and bottom pole may also be adjusted for vertical and horizontal field errors. Usually, undulators are assembled with a collection of periodic blocks surveyed to assess their accurate positions. The final process of fine-tuning the undulators requires the magnetic measurements of the whole assembly.","PeriodicalId":6760,"journal":{"name":"2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC)","volume":"18 1","pages":"1-7"},"PeriodicalIF":0.0,"publicationDate":"2020-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82514179","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 : 2020-10-31DOI: 10.1109/NSS/MIC42677.2020.9507895
K. Ali, H. Ohgaki, H. Zen, T. Kii, T. Hayakawa, T. Shizuma, H. Toyokawa, Y. Taira, M. Fujimoto, M. Katoh
We proposed an isotope-selective computed tomography (CT) imaging based on the Nuclear Resonance Fluorescence (NRF) transmission method using a quasi-monochromatic laser Compton scattering (LCS) gamma-ray beam in the MeV region for nuclear safety applications. As the first step, a two-dimensional (2D) NRF-CT image of 208Pb isotope distribution was selectively obtained for the sample target containing two enriched lead isotope rods (206Pb and 208Pb). We are planning to perform an experiment to obtain a three-dimensional NRF Computed Tomography (3D NRF-CT) image for the specific isotope. An automatic measurement system has been developed. As the result, we obtained an excellent quality of 3D gamma-ray CT image.
{"title":"Experimental Study on 3-D Isotope-Selective CT Imaging Based on Nuclear Resonance Fluorescence Transmission Method","authors":"K. Ali, H. Ohgaki, H. Zen, T. Kii, T. Hayakawa, T. Shizuma, H. Toyokawa, Y. Taira, M. Fujimoto, M. Katoh","doi":"10.1109/NSS/MIC42677.2020.9507895","DOIUrl":"https://doi.org/10.1109/NSS/MIC42677.2020.9507895","url":null,"abstract":"We proposed an isotope-selective computed tomography (CT) imaging based on the Nuclear Resonance Fluorescence (NRF) transmission method using a quasi-monochromatic laser Compton scattering (LCS) gamma-ray beam in the MeV region for nuclear safety applications. As the first step, a two-dimensional (2D) NRF-CT image of 208Pb isotope distribution was selectively obtained for the sample target containing two enriched lead isotope rods (206Pb and 208Pb). We are planning to perform an experiment to obtain a three-dimensional NRF Computed Tomography (3D NRF-CT) image for the specific isotope. An automatic measurement system has been developed. As the result, we obtained an excellent quality of 3D gamma-ray CT image.","PeriodicalId":6760,"journal":{"name":"2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC)","volume":"20 1","pages":"1-2"},"PeriodicalIF":0.0,"publicationDate":"2020-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83364933","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 : 2020-10-31DOI: 10.1109/NSS/MIC42677.2020.9507780
Guillaume Corda-D’Incan, J. Schnabel, A. Reader
We present here an enhanced version of FBSEM-Net, a deep learned regularised model-based image reconstruction algorithm. FBSEM-Net unrolls the maximum a posteriori expectation-maximisation algorithm and replaces the regularisation step by a residual convolutional neural network. Both the gradient of the prior and the regularisation strength are learnt by the network from training data. Nonetheless, some issues arise from its original implementation that we improve upon in this work to obtain a more practical implementation. Specifically, in this implementation, two theoretical improvements are included: i) iteration-dependent networks are used which allows adaptation to varying noise levels as the number of iterations evolves, ii) iteration-dependent targets are used, so that the deep learnt regulariser remains a pure denoising step without any artificial acceleration of the algorithm. Furthermore, we present a new sequential training method for fully unrolled deep networks where the iterative reconstruction is split and the network is trained on each of its modules separately to match the total number of iterations used to reconstruct the targets. The results obtained on 2D simulated test data show that FBSEM-Net using iteration-dependent networks outperforms the original version. Additionally, we found that using iteration-dependent targets not only helps to reduce the variance for different training runs of the network, thus offering greater stability, but also gives the possibility of using a lower number of iterations for test time than what was used for training. Ultimately, we demonstrate that sequential training successfully addresses potential memory issues raised during the training of unrolled networks, without notably impacting the network's performance compared to conventional training.
{"title":"Iteration-Dependent Networks and Losses for Unrolled Deep Learned FBSEM PET Image Reconstruction","authors":"Guillaume Corda-D’Incan, J. Schnabel, A. Reader","doi":"10.1109/NSS/MIC42677.2020.9507780","DOIUrl":"https://doi.org/10.1109/NSS/MIC42677.2020.9507780","url":null,"abstract":"We present here an enhanced version of FBSEM-Net, a deep learned regularised model-based image reconstruction algorithm. FBSEM-Net unrolls the maximum a posteriori expectation-maximisation algorithm and replaces the regularisation step by a residual convolutional neural network. Both the gradient of the prior and the regularisation strength are learnt by the network from training data. Nonetheless, some issues arise from its original implementation that we improve upon in this work to obtain a more practical implementation. Specifically, in this implementation, two theoretical improvements are included: i) iteration-dependent networks are used which allows adaptation to varying noise levels as the number of iterations evolves, ii) iteration-dependent targets are used, so that the deep learnt regulariser remains a pure denoising step without any artificial acceleration of the algorithm. Furthermore, we present a new sequential training method for fully unrolled deep networks where the iterative reconstruction is split and the network is trained on each of its modules separately to match the total number of iterations used to reconstruct the targets. The results obtained on 2D simulated test data show that FBSEM-Net using iteration-dependent networks outperforms the original version. Additionally, we found that using iteration-dependent targets not only helps to reduce the variance for different training runs of the network, thus offering greater stability, but also gives the possibility of using a lower number of iterations for test time than what was used for training. Ultimately, we demonstrate that sequential training successfully addresses potential memory issues raised during the training of unrolled networks, without notably impacting the network's performance compared to conventional training.","PeriodicalId":6760,"journal":{"name":"2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC)","volume":"2 1","pages":"1-4"},"PeriodicalIF":0.0,"publicationDate":"2020-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90850760","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 : 2020-10-31DOI: 10.1109/NSS/MIC42677.2020.9507973
Suzanne N. Nowicki, S. Festal, S. Czarnecki, C. Hardgrove, P. Gasda
The primary objective of the Dynamic Albedo of Neutrons (DAN) experiment on board the Mars Science Laboratory (MSL) rover Curiosity is to assess the hydrogen content as the rover traverses the Martian surface. Because hydrogen is a light element, it is an efficient moderator for neutrons. The method used to estimate the hydrogen content by the DAN instrument is to measure the thermal neutron count rate emitted from the surface of the soil using a Pulsed Neutron Generator as an activation source coupled with a thermal neutron detector. However, boron has a high cross section for thermal neutron capture and can affect the thermal neutron flux measured by the DAN instrument. Recently, the MSL ChemCam instrument has shown high concentrations of B in the veins of the Murray formation and Yellowknife Bay at concentrations of 100 to 500 ppm. We show that the number of neutrons that are captured in the Martian soil increases with increasing B, resulting in reduced count rates observed by the DAN thermal neutron detector, which can lead to an overestimate of the hydrogen content.
{"title":"The Effect of Boron on Active Neutron Measurements: Application for the Mars Science Laboratory Dynamic Albedo of Neutrons Instrument","authors":"Suzanne N. Nowicki, S. Festal, S. Czarnecki, C. Hardgrove, P. Gasda","doi":"10.1109/NSS/MIC42677.2020.9507973","DOIUrl":"https://doi.org/10.1109/NSS/MIC42677.2020.9507973","url":null,"abstract":"The primary objective of the Dynamic Albedo of Neutrons (DAN) experiment on board the Mars Science Laboratory (MSL) rover Curiosity is to assess the hydrogen content as the rover traverses the Martian surface. Because hydrogen is a light element, it is an efficient moderator for neutrons. The method used to estimate the hydrogen content by the DAN instrument is to measure the thermal neutron count rate emitted from the surface of the soil using a Pulsed Neutron Generator as an activation source coupled with a thermal neutron detector. However, boron has a high cross section for thermal neutron capture and can affect the thermal neutron flux measured by the DAN instrument. Recently, the MSL ChemCam instrument has shown high concentrations of B in the veins of the Murray formation and Yellowknife Bay at concentrations of 100 to 500 ppm. We show that the number of neutrons that are captured in the Martian soil increases with increasing B, resulting in reduced count rates observed by the DAN thermal neutron detector, which can lead to an overestimate of the hydrogen content.","PeriodicalId":6760,"journal":{"name":"2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC)","volume":"2 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2020-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88031024","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 : 2020-10-31DOI: 10.1109/NSS/MIC42677.2020.9507869
H. Arahmane, Y. Ben Maissa, E. Hamzaoui, R. E. El Moursli, J. Dumazert, A. Mahmoudi
In order to perform a fast and accurate neutron-gamma discrimination, we present in this paper a method based on supervised and unsupervised machine learning that is composed of the following steps. Firstly, we apply nonnegative parallel factor analysis to recover the original sources from mixed signals recorded at the output of a stilbene scintillator detector (45×45 mm). Secondly, spectral analysis based on the continuous wavelet transform is used to characterize these recovered original sources. Thirdly, the resulting time-scale representations are considered as images that are processed using the Otsu segmentation method in order to get the binary images and thus extract attributes of interest of neutrons and gamma-rays signals from its background. Finally, we used principal component analysis to select the most significant of these attributes that are used as inputs of a support vector machines (SVM) to discriminate and classify the neutrons from gamma-rays. To evaluate the performance of the SVM model, bias-variance analysis is used. The results show that the proposed method can achieve an operational SVM prediction model for neutron-gamma classification with a high true positive rate.
{"title":"Parallel Factor Analysis and Support Vector Machines for Neutron-Gamma Discrimination","authors":"H. Arahmane, Y. Ben Maissa, E. Hamzaoui, R. E. El Moursli, J. Dumazert, A. Mahmoudi","doi":"10.1109/NSS/MIC42677.2020.9507869","DOIUrl":"https://doi.org/10.1109/NSS/MIC42677.2020.9507869","url":null,"abstract":"In order to perform a fast and accurate neutron-gamma discrimination, we present in this paper a method based on supervised and unsupervised machine learning that is composed of the following steps. Firstly, we apply nonnegative parallel factor analysis to recover the original sources from mixed signals recorded at the output of a stilbene scintillator detector (45×45 mm). Secondly, spectral analysis based on the continuous wavelet transform is used to characterize these recovered original sources. Thirdly, the resulting time-scale representations are considered as images that are processed using the Otsu segmentation method in order to get the binary images and thus extract attributes of interest of neutrons and gamma-rays signals from its background. Finally, we used principal component analysis to select the most significant of these attributes that are used as inputs of a support vector machines (SVM) to discriminate and classify the neutrons from gamma-rays. To evaluate the performance of the SVM model, bias-variance analysis is used. The results show that the proposed method can achieve an operational SVM prediction model for neutron-gamma classification with a high true positive rate.","PeriodicalId":6760,"journal":{"name":"2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC)","volume":"51 1","pages":"1-3"},"PeriodicalIF":0.0,"publicationDate":"2020-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87426325","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 : 2020-10-31DOI: 10.1109/NSS/MIC42677.2020.9507847
Isaac Shiri, A. Akhavanallaf, Amirhossein Sanaat, Y. Salimi, D. Askari, Z. Mansouri, S. P. Shayesteh, M. Hasanian, K. Rezaei-Kalantari, A. Salahshour, S. Sandoughdaran, H. Abdollahi, H. Arabi, H. Zaidi
Chest computed tomography (CT) imaging was widely used for diagnosis and staging of severe acute respiratory syndrome coronavirus (SARS-CoV-2). CT can be utilized for initial diagnosis, severity scoring, serial monitoring, and patient status follow-up. For serial monitoring and follow-up, patients need to be scanned multiple times. The tendency in CT imaging is to minimize patient radiation dose. However, CT imaging is still considered as a high radiation dose modality. In this work, we proposed a deep residual neural network-based high quality (full dose) generation from ultra low-dose CT images to decrease the radiation dose for COVID-19 patients. In this multicenter study, we enrolled 1140 subjects with 313 PCR positive COVID-19 patients. The ultra low-dose CT images were analytically simulated, and then a deep residual neural network employed to estimate/generate full-dose images from the corresponding ultra-low-dose images. Various quantitative parameters, including the root mean square error (RMSE), structural similarity index (SSIM), and qualitative visual scoring were implemented to evaluate image quality of the generated CT images. The mean CTDIvol for full-dose images were 6.5 Gy (4.16-10.5 mGy), while, the simulated low-dose images were intended for a mean CTDIvol of 0.72 mGy (0.66-1.02 mGy). Regarding the external validation set (test set), the RMSE declined from 0.16±0.06 to 0.08±0.02 in low-dose and predicted standard-dose CT images, while the SSIM metric increased from 0.89±0.07 to 0.97±0.01, respectively. The highest visual scores (out of 5) were achieved by full-dose images (4.72±0.57) and predicted full-dose images (4.42±0.08). Conversely, ultra-low-dose images received the lowest score (2.78±0.9). In can be concluded that the proposed deep residual network improved image quality of ultra low-dose CT images, thus recovering their diagnostic value.
{"title":"Deep Residual Neural Network-based Standard CT Estimation from Ultra-Low Dose CT Imaging for COVID-19 Patients","authors":"Isaac Shiri, A. Akhavanallaf, Amirhossein Sanaat, Y. Salimi, D. Askari, Z. Mansouri, S. P. Shayesteh, M. Hasanian, K. Rezaei-Kalantari, A. Salahshour, S. Sandoughdaran, H. Abdollahi, H. Arabi, H. Zaidi","doi":"10.1109/NSS/MIC42677.2020.9507847","DOIUrl":"https://doi.org/10.1109/NSS/MIC42677.2020.9507847","url":null,"abstract":"Chest computed tomography (CT) imaging was widely used for diagnosis and staging of severe acute respiratory syndrome coronavirus (SARS-CoV-2). CT can be utilized for initial diagnosis, severity scoring, serial monitoring, and patient status follow-up. For serial monitoring and follow-up, patients need to be scanned multiple times. The tendency in CT imaging is to minimize patient radiation dose. However, CT imaging is still considered as a high radiation dose modality. In this work, we proposed a deep residual neural network-based high quality (full dose) generation from ultra low-dose CT images to decrease the radiation dose for COVID-19 patients. In this multicenter study, we enrolled 1140 subjects with 313 PCR positive COVID-19 patients. The ultra low-dose CT images were analytically simulated, and then a deep residual neural network employed to estimate/generate full-dose images from the corresponding ultra-low-dose images. Various quantitative parameters, including the root mean square error (RMSE), structural similarity index (SSIM), and qualitative visual scoring were implemented to evaluate image quality of the generated CT images. The mean CTDIvol for full-dose images were 6.5 Gy (4.16-10.5 mGy), while, the simulated low-dose images were intended for a mean CTDIvol of 0.72 mGy (0.66-1.02 mGy). Regarding the external validation set (test set), the RMSE declined from 0.16±0.06 to 0.08±0.02 in low-dose and predicted standard-dose CT images, while the SSIM metric increased from 0.89±0.07 to 0.97±0.01, respectively. The highest visual scores (out of 5) were achieved by full-dose images (4.72±0.57) and predicted full-dose images (4.42±0.08). Conversely, ultra-low-dose images received the lowest score (2.78±0.9). In can be concluded that the proposed deep residual network improved image quality of ultra low-dose CT images, thus recovering their diagnostic value.","PeriodicalId":6760,"journal":{"name":"2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC)","volume":"1 1","pages":"1-3"},"PeriodicalIF":0.0,"publicationDate":"2020-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87450784","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 : 2020-10-31DOI: 10.1109/NSS/MIC42677.2020.9507890
A. C. Whitehead, A. Biguri, N. Efthimiou, K. Su, S. Wollenweber, C. Stearns, B. Hutton, J. McClelland, K. Thielemans
Respiratory motion correction is beneficial in positron emission tomography. Different strategies for handling attenuation correction in conjunction with motion correction exist. In clinical practice, usually a single attenuation map is available, derived from computed tomography in one respiratory state. This can introduce an unwanted bias (through misaligned anatomy) into the motion correction algorithm. This paper builds upon previous work which suggested that non-attenuation corrected data was suitable for motion estimation, through the use of motion models, if time-of-flight data are available. Here, the previous work is expanded upon by incorporating attenuation correction in an iterative process. Non-attenuation corrected volumes are reconstructed using ordered subset expectation maximisation and used as input for motion model estimation. A single attenuation map is then warped to the volumes, using the motion model, the volumes are attenuation corrected, after which another motion estimation and correction cycle is performed. For validation, 4-Dimensional Extended Cardiac Torso simulations are used, for one bed position, with a field of view including the base of the lungs and the diaphragm. The output from the proposed method is evaluated against a non-motion corrected reconstruction of the same data visually, using a profile as well as standardised uptake value analysis. Results indicate that motion correction of inter-respiratory cycle motion is possible with this method, while accounting for attenuation deformation.
{"title":"PET/CT Respiratory Motion Correction With a Single Attenuation Map Using NAC Derived Deformation Fields","authors":"A. C. Whitehead, A. Biguri, N. Efthimiou, K. Su, S. Wollenweber, C. Stearns, B. Hutton, J. McClelland, K. Thielemans","doi":"10.1109/NSS/MIC42677.2020.9507890","DOIUrl":"https://doi.org/10.1109/NSS/MIC42677.2020.9507890","url":null,"abstract":"Respiratory motion correction is beneficial in positron emission tomography. Different strategies for handling attenuation correction in conjunction with motion correction exist. In clinical practice, usually a single attenuation map is available, derived from computed tomography in one respiratory state. This can introduce an unwanted bias (through misaligned anatomy) into the motion correction algorithm. This paper builds upon previous work which suggested that non-attenuation corrected data was suitable for motion estimation, through the use of motion models, if time-of-flight data are available. Here, the previous work is expanded upon by incorporating attenuation correction in an iterative process. Non-attenuation corrected volumes are reconstructed using ordered subset expectation maximisation and used as input for motion model estimation. A single attenuation map is then warped to the volumes, using the motion model, the volumes are attenuation corrected, after which another motion estimation and correction cycle is performed. For validation, 4-Dimensional Extended Cardiac Torso simulations are used, for one bed position, with a field of view including the base of the lungs and the diaphragm. The output from the proposed method is evaluated against a non-motion corrected reconstruction of the same data visually, using a profile as well as standardised uptake value analysis. Results indicate that motion correction of inter-respiratory cycle motion is possible with this method, while accounting for attenuation deformation.","PeriodicalId":6760,"journal":{"name":"2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC)","volume":"208 1","pages":"1-3"},"PeriodicalIF":0.0,"publicationDate":"2020-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76175994","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 : 2020-10-31DOI: 10.1109/NSS/MIC42677.2020.9508070
F. Mingrone, T. Pochet, Efrain Rodriguez Trujillo, Mark L. Ruch
This paper explores the possibility of using a low-noise, fast, current-sensitive preamplifier as front-end electronics for fission chambers. Over the years, these detectors have been widely used for neutron measurements in Safeguards applications, including in Unattended Monitoring Systems (UMS), which are permanently installed in nuclear facilities to continuously measure a variety of processes throughout the nuclear fuel cycle. The charge-sensitive electronics coupled to these detectors needs to be located close to the detector itself, in order to minimize the susceptibility to the noise that would arise from the capacitance of long cables. However, these devices are liable to increased failure rates when applied to harsh environments such as high neutron and gamma fields and operation at extreme temperatures. Furthermore, performing maintenance and replacing preamplifiers under these harsh conditions, or in contamination areas, can be challenging and time consuming. The usage of fast current-sensitive preamplifiers has the potential to alleviate these issues. Thanks to their high tolerance to input capacitance, they can be operated with long cables and placed in less challenging environments that are more easily accessible. In addition, being by concept much faster than charge-sensitive preamplifier, current-sensitive electronics can help when high count rates need to be measured, such as in spent fuel applications. The paper will focus on the counting performance for different cable lengths of a current-sensitive preamplifier coupled to a fission chamber, investigating possible improvements for new UMS installations.
{"title":"Evaluation of a fast current-preamplifier for use in thermal neutron detection","authors":"F. Mingrone, T. Pochet, Efrain Rodriguez Trujillo, Mark L. Ruch","doi":"10.1109/NSS/MIC42677.2020.9508070","DOIUrl":"https://doi.org/10.1109/NSS/MIC42677.2020.9508070","url":null,"abstract":"This paper explores the possibility of using a low-noise, fast, current-sensitive preamplifier as front-end electronics for fission chambers. Over the years, these detectors have been widely used for neutron measurements in Safeguards applications, including in Unattended Monitoring Systems (UMS), which are permanently installed in nuclear facilities to continuously measure a variety of processes throughout the nuclear fuel cycle. The charge-sensitive electronics coupled to these detectors needs to be located close to the detector itself, in order to minimize the susceptibility to the noise that would arise from the capacitance of long cables. However, these devices are liable to increased failure rates when applied to harsh environments such as high neutron and gamma fields and operation at extreme temperatures. Furthermore, performing maintenance and replacing preamplifiers under these harsh conditions, or in contamination areas, can be challenging and time consuming. The usage of fast current-sensitive preamplifiers has the potential to alleviate these issues. Thanks to their high tolerance to input capacitance, they can be operated with long cables and placed in less challenging environments that are more easily accessible. In addition, being by concept much faster than charge-sensitive preamplifier, current-sensitive electronics can help when high count rates need to be measured, such as in spent fuel applications. The paper will focus on the counting performance for different cable lengths of a current-sensitive preamplifier coupled to a fission chamber, investigating possible improvements for new UMS installations.","PeriodicalId":6760,"journal":{"name":"2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC)","volume":"6 1","pages":"1-3"},"PeriodicalIF":0.0,"publicationDate":"2020-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80104010","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}
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：