Trevor J. Stocki

Development and characterization of an in vitro alpha radiation exposure system

A simple in vitro alpha radiation exposure system (ARES) was designed to study the biological effects of alpha particle radiation. The ARES consists of six (241)Am electroplated stainless steel discs with activities averaging 66 kBq and Mylar-based culture dishes to allow the transmission of alpha particles. The dosimetry of the exposure system was calculated using the GEANT4 Monte Carlo simulation toolkit with the source code adapted from the open-source Microbeam example. The average dose rate and linear energy transfer of the system was simulated to be 0.98 ± 0.01 (statistical)(+0.18)( - 0.00) (systematic) Gy h(-1) and 127.4 ± 0.4 (statistical)(+23)( - 0) (systematic) keV µm(-1), respectively. The system was characterized by a comparison of the survival curves of gamma and alpha irradiated cell lines which showed a relative biological effectiveness of 6.3. This is in good agreement with values obtained using other published alpha particle exposure systems. Results show that the ARES provides a simple, cost-effective exposure platform for research into the biological effects of alpha particle radiation using in vitro modelling of cell cultures.

Cosmic Ray Inspection and Passive Tomography for SNM Detection

The Cosmic Ray Inspection and Passive Tomography (CRIPT) project has recently started investigating the detection of illicit Special Nuclear Material in cargo using cosmic ray muon tomography and complementary neutron detectors. We are currently performing simulation studies to help with the design of small scale prototypes. Based on the prototype tests and refined simulations, we will determine whether the muon tracking system for the full scale prototype will be based on drift chambers or extruded scintillator trackers. An analysis of the operations of the Port of Montreal has determined how long muon scan times should take if all or a subset of the cargo is to be screened. As long as the throughput of the muon system(s) is equal to the rate at which containers are unloaded from ships, the impact on port operations would not be great if a muon scanning stage were required for all cargo. We also show preliminary simulation results indicating that excellent separation between Al, Fe and Pb is possible under ideal conditions. The discrimination power is reduced but still significant when realistic momentum resolution measurements are considered.

Full border identification for reduction of training sets

Border identification (BI) was previously proposed to help learning systems focus on the most relevant portion of the training set so as to improve learning accuracy. This paper argues that the traditional BI implementation suffers from a serious limitation: it is only able to identify partial borders. This paper proposes a new BI method called Progressive Border Sampling (PBS), which addresses this limitation by borrowing ideas from recent research on Progressive Sampling. PBS progressively learns optimal borders from the entire training sets by, first, identifying a full border, thus, avoiding the limitation of the traditional BI method, and, second, by incrementing the size of that border until it converges to an optimal sample, which is smaller than the original training set. Since PBS identifies the full border, it is expected to discover more optimal samples than traditional BI. Our experimental results on the selected 30 benchmark datasets from the UCI repository show that, indeed, in the context of classification, PBS is more successful than traditional BI at reducing the size of the training sets and optimizing the accuracy results.

Effects of alpha particle radiation on gene expression in human pulmonary epithelial cells

The general public receives approximately half of its exposure to natural radiation through alpha (α)-particles from radon ((222)Rn) gas and its decay progeny. Epidemiological studies have found a positive correlation between exposure to (222)Rn and lung carcinogenesis. An understanding of the transcriptional responses involved in these effects remains limited. In this study, genomic technology was employed to mine for subtle changes in gene expression that may be representative of an altered physiological state. Human lung epithelial cells were exposed to 0, 0.03, 0.3 and 0.9Gy of α-particle radiation. Microarray analysis was employed to determine transcript expression levels 4h and 24h after exposure. A total of 590 genes were shown to be differentially expressed in the α-particle radiated samples (false discovery rate (FDR)≤0.05). Sub-set of these transcripts were time-responsive, dose-responsive and both time- and dose-responsive. Pathway analysis showed functions related to cell cycle arrest, and DNA replication, recombination and repair (FDR≤0.05). The canonical pathways associated with these genes were in relation to pyrimidine metabolism, G2/M damage checkpoint regulation and p53 signaling (FDR≤0.05). Overall, this gene expression profile suggests that α-particle radiation inhibits DNA synthesis and subsequent mitosis, and causes cell cycle arrest.

{gamma} rays from muon capture in I, Au, and Bi

A significant improvement has been made in the identification of {gamma} rays from muon capture in I, Au, and Bi, all monisotopic elements. The ({mu}{sup -},{nu}n) reaction was clearly observed in all nuclei, but the levels excited do not correlate well with the spectroscopic factors from the (d,{sup 3}He) reaction. Some ({mu}{sup -},{nu}2n), ({mu}{sup -},{nu}3n), ({mu}{sup -},{nu}4n), ({mu}{sup -},{nu}5n) and other reactions have been observed at a lower yield. The muonic x-ray cascades have also been studied in detail.

Machine learning for radioxenon event classification for the Comprehensive Nuclear-Test-Ban Treaty.

A method of weapon detection for the Comprehensive nuclear-Test-Ban-Treaty (CTBT) consists of monitoring the amount of radioxenon in the atmosphere by measuring and sampling the activity concentration of (131m)Xe, (133)Xe, (133m)Xe, and (135)Xe by radionuclide monitoring. Several explosion samples were simulated based on real data since the measured data of this type is quite rare. These data sets consisted of different circumstances of a nuclear explosion, and are used as training data sets to establish an effective classification model employing state-of-the-art technologies in machine learning. A study was conducted involving classic induction algorithms in machine learning including Naïve Bayes, Neural Networks, Decision Trees, k-Nearest Neighbors, and Support Vector Machines, that revealed that they can successfully be used in this practical application. In particular, our studies show that many induction algorithms in machine learning outperform a simple linear discriminator when a signal is found in a high radioxenon background environment.

Instance Selection by Border Sampling in Multi-class Domains

Instance selection is a pre-processing technique for machine learning and data mining. The main problem is that previous approaches still suffer from the difficulty to produce effective samples for training classifiers. In recent research, a new sampling technique, called Progressive Border Sampling (PBS), has been proposed to produce a small sample from the original labelled training set by identifying and augmenting border points. However, border sampling on multi-class domains is not a trivial issue. Training sets contain much redundancy and noise in practical applications. In this work, we discuss several issues related to PBS and show that PBS can be used to produce effective samples by removing redundancies and noise from training sets for training classifiers. We compare this new technique with previous instance selection techniques for learning classifiers, especially, for learning Naive Bayes-like classifiers, on multi-class domains except for one binary case which was for a practical application.

Border Sampling through Coupling Markov Chain Monte Carlo

Recently, progressive border sampling (PBS) was proposed for sample selection in supervised learning by progressively learning an augmented full border from small labeled datasets. However, this quadratic learning algorithm is inapplicable to large datasets. In this paper, we incorporate the PBS to a state of the art technique called coupling Markov chain Monte Carlo (CMCMC) in an attempt to scale the original algorithm up on large labeled datasets. The CMCMC can produce an exact sample while a naive strategy for Markov chain Monte Carlo cannot guarantee the convergence to a stationary distribution. The resulting CMCMC-PBS algorithm is thus proposed for border sampling on large datasets. CMCMC-PBS exhibits several remarkable characteristics: linear time complexity, learner-independence, and a consistent convergence to an optimal sample from the original training sets by learning from their subsamples. Our experimental results on the 33 either small or large labeled datasets from the UCIKDD repository and a nuclear security application show that our new approach outperforms many previous sampling techniques for sample selection.

Detection of anthropogenic radionuclides by the CA002 monitoring station for the comprehensive test ban treaty.

A worldwide monitoring system for radioactive aerosols is being implemented for verification of the Comprehensive Test Ban Treaty. These 80 stations will detect airborne radioactivity not only from nuclear explosions but also from other anthropogenic and natural sources. A prototype unit has been in operation since April 1996 in Vancouver, British Columbia, Canada. It is a very sensitive system and reports clear signals for natural radioactivity, including cosmogenic 7Be, and the decay products from soil exhalation of 220Rn (thoron). In addition, there have been frequent detections of anthropogenic nuclides, probably coming from three distinct facilities-a medical isotope production center, a major university hospital, and a particle accelerator laboratory--all between 1 and 2 km away from the monitoring station. This experience is discussed to sensitize health physicists to the potential uses of this publicly available information.

FIRST IMAGES FROM THE CRIPT MUON TOMOGRAPHY SYSTEM

The CRIPT Cosmic Ray Imaging and Passive Tomography system began data taking in September 2012. CRIPT is a “proof of principle” muon tomography system originally proposed to inspect cargo in shipping containers and to determine the presence of special nuclear materials. CRIPT uses 4 layers of 2 m x 2 m scintillation counter trackers, each layer measuring two coordinates. Two layers are used to track the incoming muon and two for the outgoing muon allowing the trajectories of the muon to be determined. The target volume is divided into voxels, and a Point of Closest Approach algorithm is used to determine the number of scattering events in each voxel, producing a 3D image. The system has been tested with various targets of depleted uranium, lead bricks, and tungsten rods. Data on the positional resolution has been taken and the intrinsic resolution is unfolded with the help of a simulation using GEANT4. The next steps include incorporation of data from the spectrometer section, which will assist in determining the muons momentum and improve the determination of the density of the target.