Comparison of active interrogation methods for source location in a scattering and absorbing medium, consisting of PGNAA, and an AmBe quasi-forward biased directional source.

Source location of Special Nuclear Material (SNM) encompassing 95% 235U and 239Pu is identified by utilizing a directional source from Patent No: US20190013109A1 using Prompt Gamma Neutron Activation Analysis (PGNAA) and neutron spectroscopy simulated with Monte-Carlo N-Particle transport 6.2 (MCNP). BC-408, HPGe, LaBr3 detector arrays were used to identify the location of the SNM using total counts incident on each detector, and PGNAA photopeaks from HPGe and LaBr3 detector arrays in a polyethylene shield. The conducted simulations varied the volume and location of the SNM in the MCNP input files to observe how the source location method behaved. PGNAA photopeaks used for source identification include 61 keV from fission, 2.223 MeV prompt gamma from hydrogen, 511 keV annihilation, and a single and double escape peaks from the prompt gamma interaction from hydrogen. The capabilities of each detector systems to acquire well resolved photopeaks with a 1% relative error or less, and total relative error for F4 and F8 tallies were less than 0.015% relative error. Source predictions of the SNM with uneven amounts of polyethylene shielding between the source and detectors was observed to overpredict and give invalid source location predictions. Source locations of the SNM with even amounts of polyethylene material between the source and each detector were found to be valid. With a 1 Ci 241Am source activity, it was determined that 1630 s were needed to obtain the results for each detector system with the quasi-forward directional AmBe source. Coupling source and material identification together would increase acquisition time but would only require one system to determine.

Looking Inside a Prototype Compact X-Ray Tube Comprising CNT-Based Cold Cathode and Transmission-Type Anode.

In this article, we offer a look inside our prototype compact X-ray tube by reporting on our findings when we experimentally studied it. We studied the prototype experimentally to characterize its primary components, i.e., carbon nanotube (CNT)-based cold cathode, electrostatic lens and transmission-type anode, and to validate our previous simulation studies. We characterized the CNT-based cold cathode by studying the relationship between the electron emission current and its control parameter, electron extraction voltage. This relationship, commonly known as the current-voltage characteristic, showed an exponential-like nature that is expected from the Fowler-Nordheim model for field emission. Next, we characterized the electrostatic lens by studying the effects of lens voltage on the focal spot size. Their relationship showed a "V" trend and corroborated that we can control the focal spot size by controlling the lens voltage. We then characterized the transmission-type anode of the prototype by studying its output X-ray energy spectra at different anode voltages. We could control the highest and the mean X-ray energies generated from the transmission-type anode using the anode voltage. For the same anode voltage and aluminum filtration, when we compared the Xray energy spectrum generated from the transmission-type anode with that of the conventional reflection-type anode, we observed that the two energy spectra agreed with each other.

Active interrogation of Special Nuclear Material containers using AmBe quasi-forward biased directional source and PGNAA.

Special Nuclear Material (SNM) including 95% 235U and 239Pu is capable of being identified by utilizing a quasi-forward directional AmBe source from Patent No: US20190013109A1 using Prompt Gamma Neutron Activation Analysis (PGNAA) simulated with Monte-Carlo N-Particle transport 6.2 (MCNP). HPGe and LaBr3 detector arrays were used to identify and quantify the peak-to-background and peak-to-total ratios of the associated photon spectra from the SNM encased in a polyethylene shield. The conducted simulations varied the volume of the SNM and neutron source strength in the MCNP data card to conduct an uncertainty analysis. Photopeaks identified include K-shell X-rays from 235U and 239Pu, 61 keV from fission, 2.223 MeV prompt gamma from hydrogen, 511 keV annihilation, and a single and double escape peak from the prompt gamma interaction from hydrogen. Relationships between peak-to-total and peak-to-background as a function of SNM and particle history were investigated to aid in the analysis. The capabilities of both detector systems to acquire well resolved photopeak with a 5% relative error or less, with a 1 Ci source activity, and a peak-to-background ratio of 1.15. This was determined to take 326 s for LaBr3 and 163 s for HPGe which is comparable to current methods for material detection which take between 100 and 900 s to acquire.

A fast and scalable method for quality assurance of deformable image registration on lung CT scans using convolutional neural networks.

PURPOSE: To develop and evaluate a method to automatically identify and quantify deformable image registration (DIR) errors between lung computed tomography (CT) scans for quality assurance (QA) purposes. METHODS: We propose a deep learning method to flag registration errors. The method involves preparation of a dataset for machine learning model training and testing, design of a three-dimensional (3D) convolutional neural network architecture that classifies registrations into good or poor classes, and evaluation of a metric called registration error index (REI) which provides a quantitative measure of registration error. RESULTS: Our study shows that, despite having limited number of training images available (10 CT scan pairs for training and 17 CT scan pairs for testing), the method achieves 0.882 AUC-ROC on the test dataset. Furthermore, the combined standard uncertainty of the estimated REI by our model lies within ± 0.11 (± 11% of true REI value), with a confidence level of approximately 68%. CONCLUSIONS: We have developed and evaluated our method using original clinical registrations without generating any synthetic/simulated data. Moreover, test data were acquired from a different environment than that of training data, so that the method was validated robustly. The results of this study showed that our algorithm performs reasonably well in challenging scenarios.

A Monte Carlo simulation study of a flat-panel X-ray source.

A novel flat-panel transmission X-ray source based on nitrogen-incorporated ultra-nano-crystalline diamond (N-UNCD) field emitters is being developed for medical and industrial X-ray imaging. X-ray generation characteristics of the X-ray tubes were simulated with the MCNPX 2.6.0 Monte Carlo code. The simulation results showed that 0.25-12 µm thick tungsten targets could generate an adequate intensity of Bremsstrahlung X-rays. Generated X-ray output intensities of 1.48×10(12)MeV/mA-s of 100 kVp electrons hitting the target were found before collimation.