Abstract Aims To develop an artificial intelligence (AI)-model which enables fully automated accurate quantification of coronary artery calcium (CAC), using deep learning (DL) on electrocardiogram (ECG)-gated non-contrast cardiac computed tomography (gated CCT) images. Methods and results Retrospectively, 560 gated CCT images (including 60 synthetic images) performed at our institution were used to train AI-model, which can automatically divide heart region into five areas belonging to left main (LM), left anterior descending (LAD), circumflex (LCX), right coronary artery (RCA), and another. Total and vessel-specific CAC score (CACS) in each scan were manually evaluated. AI-model was trained with novel Heart-labelling method via DL according to the manual-derived results. Then, another 409 gated CCT images obtained in our institution were used for model validation. The performance of present AI-model was tested using another external cohort of 400 gated CCT images of Stanford Center for Artificial Intelligence of Medical Imaging by comparing with the ground truth. The overall accuracy of the AI-model for total CACS classification was excellent with Cohen’s kappa of k = 0.89 and 0.95 (validation and test, respectively), which surpasses previous research of k = 0.89. Bland-Altman analysis showed little difference in individual total and vessel-specific CACS between AI-derived CACS and ground truth in test cohort (mean difference [95% confidence interval] were 1.5 [−42.6, 45.6], −1.5 [−100.5, 97.5], 6.6 [−60.2, 73.5], 0.96 [−59.2, 61.1], and 7.6 [−134.1, 149.2] for LM, LAD, LCX, RCA, and total CACS, respectively). Conclusion Present Heart-labelling method provides a further improvement in fully automated, total, and vessel-specific CAC quantification on gated CCT.

ObjectiveIn this study, we investigated the optimal reconstruction algorithm in fluorodeoxyglucose (FDG) positron emission tomography (PET) with a short acquisition time.Materials and MethodsIn the phantom study, six spheres filled with FDG solution (sphere size: 6.23-37 mm; radioactivity ratio of spheres to background = 8:1) and placed in a National Electrical Manufacturers Association phantom were evaluated. Image acquisition time was 15 to 180 seconds, and the obtained image data were reconstructed using each of the Fourier rebinning (FORE) + ordered subsets expectation-maximization (OSEM) and 3D-OSEM algorithms. In the clinical study, mid-abdominal images of 19 patients were evaluated using regions of interest placed on areas of low, intermediate, and high radioactivity. All obtained images were investigated visually, and quantitatively using maximum standardized uptake value (SUV) and coefficient of variation (CV).ResultsIn the phantom study, FORE + OSEM images with a short acquisition time had large CVs (poor image quality) but comparatively constant maximum SUVs. 3D-OSEM images showed comparatively constant CVs (good image quality) but significantly low maximum SUVs. The results of visual evaluation were well correlated with those of quantitative evaluation. Small spheres were obscured on 3D-OSEM images with short acquisition time, but image quality was not greatly deteriorated. The clinical and phantom studies yielded similar results.ConclusionFDG PET images with a short acquisition time reconstructed by FORE + OSEM showed poorer image quality than by 3D-OSEM. However, images obtained with a short acquisition time and reconstructed with FORE + OSEM showed clearer FDG uptake and more useful than 3D-OSEM in the light of the detection of lesions.

We aimed to examine the possibility that polyethylene glycol (PEG) phantoms can simulate apparent diffusion coefficients (ADCs) of malignant tumors and the effectiveness of PEG phantoms using three 3.0 T magnetic resonance imaging (MRI) systems. In particular, the correlations between PEG concentrations and ADC values, the validation of ADC measurement precision, and the stability and reproducibility of PEG phantom were verified. A phantom containing 0, 0.625, 1.250, 2.5, 5, 10, 20 mM PEG was assessed using three MRI systems. The endpoints comprised correlations between PEG concentrations and ADC, validation of ADC measurement precision and the stability and reproducibility of the PEG phantom. The correlation coefficients between PEG concentrations and the ADC of the three MRI systems and among the three MRI systems revealed negative (r ≈ − 1.000, P < 0.001) and positive (r ≈ 1.000, P < 0.001) correlations. The ADCmean of 2.5–20 mM PEG was significant (P < 0.05 t tests), and that at ≥ 10 mM PEG was < 1.0 × 10–3 mm2/s. The %CV ranged from 1.20 to 4.62, and repeatability was confirmed by the 90% confidence interval. The maximum values for DifferenceChange over time and DifferencePri.vs.New were 0.20 and 0.19 × 10–3 mm2/s, respectively. In this study, we found that PEG concentrations ≥ 10 mM are required to simulate the ADCs of malignant tumors (ADC < 1.0 × 10–3 mm2/s). We also showed that the ADC value is easily controlled by adjusting PEG concentrations, and can be stably measured using our PEG phantom for at least 6 months. The PEG phantom can easily and stably simulate the ADC of malignant tumors with high reproducibility.

Purpose: The purpose of this study is to evaluate the effects of cone-beam computed tomography (CBCT) on dose distribution and normal tissue complication probability (NTCP) by constructing a comprehensive dose evaluation system for prostate intensity-modulated radiation therapy (IMRT). Methods: A system that could combine CBCT and treatment doses with MATLAB was constructed. Twenty patients treated with prostate IMRT were studied. A mean dose of 78 Gy was prescribed to the prostate region, excluding the rectal volume from the target volume, with margins of 4 mm to the dorsal side of the prostate and 7 mm to the entire circumference. CBCT and treatment doses were combined, and the dose distribution and the NTCP of the rectum and bladder were evaluated. Results: The radiation dose delivered to 2% and 98% of the target volume increased by 0.90 and 0.74 Gy on average, respectively, in the half-fan mode and on average 0.76 and 0.72 Gy, respectively, in the full-fan mode. The homogeneity index remained constant. The percent volume of the rectum and bladder irradiated at each dose increased slightly, with a maximum increase of <1%. The rectal NTCP increased by approximately 0.07% from 0.46% to 0.53% with the addition of a CBCT dose, while the maximum NTCP in the bladder was approximately 0.02%. Conclusions: This study demonstrated a method to evaluate a combined dose of CBCT and a treatment dose using the constructed system. The combined dose distribution revealed increases of <1% volume in the rectal and bladder doses and approximately 0.07% in the rectal NTCP.

Purpose: The purpose of this study is to evaluate the influence of variable relative biological effectiveness (RBE) of proton beam and dose fractionation has on dose distribution and to establish a new three-dimensional dose evaluation method for proton therapy combined with high-dose-rate (HDR) brachytherapy. Materials and Methods: To evaluate the influence of variable RBE and dose fractionation on dose distribution in proton beam therapy, the depth-dose distribution of proton therapy was compared with clinical dose, RBE-weighted dose, and equivalent dose in 2 Gy fractions using a linear-quadratic-linear model (EQD2LQL). The clinical dose was calculated by multiplying the physical dose by RBE of 1.1. The RBE-weighted dose is a biological dose that takes into account RBE variation calculated by microdosimetric kinetic model implemented in Monte Carlo code. The EQD2LQLis a biological dose that makes the RBE-weighted dose equivalent to 2 Gy using a linear-quadratic-linear (LQL) model. Finally, we evaluated the three-dimensional dose by taking into account RBE variation and LQL model for proton therapy combined with HDR brachytherapy. Results: The RBE-weighted dose increased at the distal of the spread-out Bragg peak (SOBP). With the difference in the dose fractionation taken into account, the EQD2LQLat the distal of the SOBP increased more than the RBE-weighted dose. In proton therapy combined with HDR brachytherapy, a divergence of 103% or more was observed between the conventional dose estimation method and the dose estimation method we propose. Conclusions: Our dose evaluation method can evaluate the EQD2LQLconsidering RBE changes in the dose distribution.

Objective The aim of this study was to investigate the efficacy of compressed sensing (CS)-based iterative reconstruction (CS-IR) from undersampled projection data in 123I-N-&ohgr;-fluoropropyl-2&bgr;-carbomethoxy-3&bgr;-(4-iodophenyl)nortropane single-photon emission computed tomography (SPECT). Materials and methods We used the cylinder/sphere and the striatal digital phantom models. The number of projections was set at 120, 90, 60, 40, and 30 projections. SPECT images were reconstructed using filtered back-projection (FBP), maximum likelihood-expectation maximization (ML-EM), and CS-IR. The total-variation transform with local image gradient in L1-norm was adopted in our CS algorithm. The efficacy of CS-IR was examined in terms of the spatial resolution, recovery coefficient, aspect ratio (ASR), activity concentration linearity, percent coefficient of variation (%CV), and specific binding ratio. Results As the number of projections decreased, the following results were observed. No differences of the spatial resolution and activity concentration linearity were observed between reconstruction methods. However, ASR for FBP slightly increased in contrast to ML-EM and CS-IR for which ASR remained constant. There were not any clear differences between recovery coefficients obtained from each reconstruction. The %CV obtained by CS-IR was significantly superior to that obtained by other reconstructions at all number of projections: for example, the %CV obtained by 60 projection CS-IR was equivalent to that obtained by 120 projection FBP and ML-EM. The specific binding ratio did not change with the number of projections, and there were no significant differences between FBP, ML-EM, and CS-IR. Conclusion We have demonstrated that CS-IR with decreased number of projections can provide a good image quality compared with commonly used SPECT reconstruction methods. This CS could help to reduce overall acquisition time in 123I-N-&ohgr;-fluoropropyl-2&bgr;-carbomethoxy-3&bgr;-(4-iodophenyl)nortropane SPECT, particularly.

Objective The aim of this study was to investigate the efficacy of compressed sensing (CS)-based iterative reconstruction (CS-IR) from undersampled projection data in I- 123-N-omega-fluoropropyI-2 beta-carbomethoxy-3 beta-(4-iodophenyl)nortropane single-photon emission computed tomography (SPECT).Materials and methods We used the cylinder/sphere and the striatal digital phantom models. The number of projections was set at 120, 90, 60, 40, and 30 projections. SPECT images were reconstructed using filtered back-projection (FBP), maximum likelihood-expectation maximization (ML-EM), and CS-IR. The total-variation transform with local image gradient in L1-norm was adopted in our CS algorithm. The efficacy of CS-IR was examined in terms of the spatial resolution, recovery coefficient, aspect ratio (ASR), activity concentration linearity, percent coefficient of variation (%CV), and specific binding ratio.Results As the number of projections decreased, the following results were observed. No differences of the spatial resolution and activity concentration linearity were observed between reconstruction methods. However, ASR for FBP slightly increased in contrast to ML-EM and CS-IR for which ASR remained constant. There were not any clear differences between recovery coefficients obtained from each reconstruction. The %CV obtained by CS-IR was significantly superior to that obtained by other reconstructions at all number of projections: for example, the %CV obtained by 60 projection CS-I R was equivalent to that obtained by 120 projection FBP and ML-EM. The specific binding ratio did not change with the number of projections, and there were no significant differences between FBP, ML-EM, and CS-IR.Conclusion We have demonstrated that CS-I R with decreased number of projections can provide a good image quality compared with commonly used SPECT reconstruction methods. This CS could help to reduce overall acquisition time in(123)I-N-omega-fluoropropyI-2 beta-carbomethoxy-3 beta-(4-iodophenyl)nortropane SPECT, particularly. (C) 2018 Wolters Kluwer Health, Inc. All rights reserved.

The aim of this study was to evaluate the effects of the acquisition rotation speed and the rotation time for continuous repetitive rotation acquisition (CRRA) on image quality and quantification in 123I-FP-CIT SPECT. Methods: An anthropomorphic striatal phantom filled with 123I solution was acquired with CRRA and the step-and-shoot (SS) mode. The following combinations of acquisition rotation speed and rotation time for CRRA were used: 0.50 rpm by 30 frames, 0.17 rpm by 10 frames, 0.10 rpm by 6 frames, and 0.05 rpm by 3 frames. SPECT images were reconstructed using ordered-subset expectation maximization with resolution recovery, scatter, and CT-based attenuation correction. Two kinds of image processing patterns-image reconstruction after the addition of projection data (the added-projection-data process) and image addition after data reconstruction (the added-reconstructed-image process)-were investigated in this study. The effects of the acquisition parameters and the image processes were evaluated by the full width at half maximum, percentage coefficient of variation (%CV), and specific binding ratio (SBR). Results: With full width at half maximum, there were no clear differences between CRRA images obtained with the various rotation speeds before rotation and the SS mode. Although the combination of a slow rotation speed and a short rotation time improved image uniformity compared with the SS mode, the %CV obtained by CRRA increased as the rotation speed increased. The %CVs were 11.9% ± 0.9% for 0.50 rpm by 30 frames, 6.9% ± 0.9% for 0.05 rpm by 3 frames, and 9.6% ± 0.5% for SS mode. SBRs obtained by CRRA with the added-projection-data process were equal to those obtained by SS mode. However, SBRs obtained with the added-reconstructed-image process were clearly decreased compared with the SS mode. Conclusion: The combination of rotation speed and rotation times affects the image quality and quantification of 123I-FP-CIT SPECT using CRRA. When CRRA is applied in 123I-FP-CIT SPECT, it is necessary to use added-projection-data processes and proper rotation speeds (e.g., 0.10-0.17 rpm rotation speed).

PURPOSE:The aim of this study was to evaluate the effect of misregistration between single-photon emission computed tomography (SPECT) and computed tomography (CT) images on bone SPECT.METHODS:We acquired SPECT and CT images of a body phantom filled with bone-equivalent solution and 99mTc for evaluation of bone SPECT. SPECT images were reconstructed using attenuation correction maps obtained by shifting the attenuation coefficients from non-shifted values (reference). Activity concentrations, SPECT standardized uptake values (SPECT-SUVs), and tumor background ratios (TBRs) were evaluated.RESULTS:Activity concentrations and SPECT-SUVs decreased with decreasing attenuation coefficient. The difference in attenuation coefficient was especially large between the shifted-to-lung (0.085 cm-1) and reference (0.249 cm-1) values. Non-shifted and shifted-to-lung SPECT-SUVs were 11.5±1.0 and 2.3±0.2, respectively. TBR also decreased with decreasing attenuation coefficient. The maximum percentage change in TBR was 86% in the shifted-to-lung value.CONCLUSIONS:Our results indicate that the accuracy of activity concentration and lesion detectability was commonly affected by misalignment between SPECT and CT images. Although the impact of SPECT/CT misregistration on bone SPECT is case-specific and difficult to predict, it is important to reduce the incidence of misregistration errors for quantitative bone SPECT imaging.

Specific design features and main characteristics of a compact accelerator-based epithermal neutron source for boron neutron capture therapy are described. High quality of the neutron flux generated by the accelerator-based source has been experimentally confirmed. The opportunities for medical use of the accelerator-based epithermal neutron source in oncological centers for boron neutron capture therapy are assessed.

PURPOSE:Non-uniformity of a scintillation camera can result in artifacts on planar, projection, and single-photon emission computed tomography (SPECT) images. The purpose of this study was to evaluate the effect of field uniformity on artifact generation. METHODS:Using a simulation phantom, we investigated the relationship between non-uniformity of the image and artifacts on planar, projection, and SPECT images. All the non-uniformity images were generated by decreasing the photomultiplier tube sensitivity ranging from 0% to 10%. Quantitative analysis was performed using integral and differential uniformity. We also visually assessed artifact magnitude. RESULTS:Integral and differential uniformity increased with decreasing the photomultiplier tube sensitivity and tended to be higher in SPECT images compared with planar and projection images. For visual assessment, mean scores in SPECT images were higher than in planar and projection images for artifact detection. CONCLUSIONS:Our results indicated that decreasing field uniformity is expected to produce artifacts in planar and SPECT images. Also, SPECT images require very high-field uniformity.

Purpose: The present study aimed to define the errors in SUV and demonstrate the feasibility of SUV harmonization among contemporary PET/CT scanners using a novel National Institute of Standards and Technology (NIST)-traceable Ge-68/Ga-68 source as the reference standard. Methods: We used Ge-68/Ga-68 dose calibrator and PET sources made with same batch of Ge-68/Ga-68 embedded in epoxy that is traceable to the NIST standard. Bias in the amount of radioactivity and the radioactive concentrations measured by the dose calibrators and PET/CT scanners, respectively, was determined at five Japanese sites. We adjusted optimal dial setting of the dose calibrators and PET reconstruction parameters to close the actual amount of radioactivity and the radioactive concentration, respectively, of the NIST-traceable Ge-68/Ga-68 sources to harmonize SUV. Errors in SUV before and after harmonization were then calculated at each site. Results: The average bias in the amount of radioactivity and the radioactive concentrations measured by dose calibrator and PET scanner was - 4.94% and - 12.22%, respectively, before, and - 0.14% and - 4.81%, respectively, after harmonization. Corresponding averaged errors in SUV measured under clinical conditions were underestimated by 7.66%, but improved by - 4.70% under optimal conditions. Conclusion: Our proposed method using an NIST-traceable 68Ge/68Ga source identified bias in values obtained using dose calibrators and PET scanners, and reduced SUV variability to within 5% across different models of PET scanners at five sites. Our protocol using a standard source has considerable potential for harmonizing the SUV when contemporary PET scanners are involved in multicenter studies.

In the current article, we provide in vitro efficacy evaluation of a unique accelerator-based neutron source, constructed at the Budker Institute of Nuclear Physics (Novosibirsk, Russian Federation), for boron neutron capture therapy (BNCT), which is particularly effective in the case of invasive cancers. U251MG, CHO-K1 and V79 cells were incubated and irradiated in various concentrations of boric acid with epithermal neutrons for 2-3 h in a plexiglass phantom, using 2.0 MeV proton energy and 1.5-3.0 mA proton current, resulting in a neutron fluence of 2.16 × 1012 cm-2. The survival curves of cells loaded with boron were normalized to those irradiated without boron (to exclude the influence of the fast neutron and gamma dose components) and fit to the linear-quadratic (LQ) model. Colony formation assays showed the following cell survival rates (means ± SDs): CHO-K1: 0.348 ± 0.069 (10 ppm), 0.058 ± 0.017 (20 ppm), 0.018 ± 0.005 (40 ppm); V79: 0.476 ± 0.160 (10 ppm), 0.346 ± 0.053 (20 ppm), 0.078 ± 0.015 (40 ppm); and U251MG: 0.311 ± 0.061 (10 ppm), 0.131 ± 0.022 (20 ppm), 0.020 ± 0.010 (40 ppm). The difference between treated cells and controls was significant in all cases (P < 0.01) and confirmed that the neutron source and irradiation regimen were sufficient for control over cell colony formation. We believe our study will serve as a model for ongoing in vitro experiments on neutron capture therapy to advance in this area for further development of accelerator-based BNCT into the clinical phase.

Electronic portal imaging device (EPID) is a popular tool for a verification of the patient setup for radiation therapy. Recently, it is grabbing attention as a tool for in vivo dosimetry. It detects therapeutic photons which penetrated the patient body to acquire an image. Conversion to absorbed dose is required from the digital pixel value. However, the response of the EPID is susceptible to the energy spectrum changes through the body [1] , [2] . The purpose of this study was to perform the energy response correction of the EPID image using Monte Carlo simulation (MC) to improve the accuracy of the EPID dosimetry. Geant4.9.6 patch04 was used for the MC with a combination of a HTCondor 7.6 parallel computing software. The MC simulated the 6 and 10 MV photon beams from Trilogy (Varian Medical Systems) linear accelerator. The EPID response to the energy spectrum was evaluated. CT DICOM data set of a chest phantom was imported to the MC, and the simulated EPID images were acquired after irradiating 6 and 10 MV photon beams. The measurement was performed with the same setup as the MC and obtained the EPID images. The energy response was corrected to obtain the absorbed dose and compared with that measured with 2D array detector (PTW) as a reference. EPID was more sensitive to the photons penetrated the mediastinum than those penetrated the lung due to the contamination of the scattered photons with the energy of