This PDF file contains the front matter associated with SPIE Proceedings Volume 13043, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

There is a need for novel sensors and detection systems that can be applied to multiple use cases in Homeland Security environments. These detection systems, such as those using x-ray imaging technology, should be able to detect a variety of concealed threats in multiple types of conveyances for U.S. Department of Homeland Security (DHS) components and other groups that need this capability. An example is the need for a technology that can detect explosives for Transportation Security Agency in checked baggage and can detect fentanyl for Customs and Border Protection in mail. The goal is to reduce the development and operation costs because the technologies are multi-use. Additional requirements for these technologies include faster response to emerging threats and the ability to integrate with other technologies in systems of systems. This paper will discuss current emerging technology trends in threat detection of concealed threats along with research and development approaches DHS Science & Technology Directorate is taking to explore the multi-use approach to detect not only explosives but other chemicals of interest such as illicit drugs in these complex environments.

Medical, industrial and security x-ray systems should give images that demonstrate material contrast for accurate identification. Acquiring images simultaneously from coherently scattered x-rays plus primary is an efficient way to do so. In our projection imaging system for small biological samples, the same detector is used for both primary and scatter, necessitating attenuation of the post-object primary in order that both primary and scatter lie within the detector’s dynamic range. Consideration of the cross sections shows that the peak scatter-to-primary fluence ratio is nearly independent of photon energy. At lower energies, coherent scatter produces larger diffraction rings, which give better spatial separation of primary and scatter, but there is more attenuation and multibeam information disentanglement is more difficult. Previous development work for our system used a 110kV incident beam, with a 1.5-mm-thick post-object attenuator disk of 90% W/10% Cu alloy for each of the 15 pencil beams. In this work we investigate alternatives which are less attenuating. By reducing the kV and using K-edge filters the beam average energy is lowered to achieve greater primary image contrast. Preliminary primary beam results are shown.

The advent of photon-counting detectors (PCDs) promises a significant shift in clinical x-ray imaging by capturing individual x-ray photon's intensity and energy, unlike traditional detectors. This enables energy discrimination, which has the potential to improve image quality, reduce radiation dose, and enable new diagnostic applications. PCDs can significantly improve spectral x-ray imaging systems for tissue characterization and material separation, paving the way for CT color and quantitative imaging. Most major CT companies have recently been developing prototype PCD CT systems. Two CT systems incorporating a PCD were FDA cleared within the last two years. Furthermore, PCDs empower novel acquisition techniques beyond conventional x-ray transmission, including coherent scatter and phase contrast imaging. This abstract outlines key advancements in spectral imaging via PCDs, emphasizing studies and assessments conducted at the FDA's Office of Science and Engineering Laboratories (OSEL) and Division of Imaging, Diagnostics, and Software Reliability (DIDSR). The focus is on the transformative potential of PCD-equipped spectral x-ray systems in enhancing diagnostic imaging.

Silicon field emitter arrays are a promising new cold-cathode technology which can offer high absolute currents, current density and excellent temporal resolution. Progress in developing silicon field emitter arrays as a platform technology to demonstrate compact, high-performance and inexpensive x-ray sources will be discussed in this presentation.

Identifying and intercepting prohibited items and explosives is a critical focus of aviation security. While computed tomography (CT) systems represent the industry standard for detecting explosives in baggage, x-ray diffraction imaging (XRDI) systems have shown increasing performance and commercial viability. Our approach to explosives detection involves the combination of CT and XRDI into a single, hybrid system where both the CT and XRDI data are utilized in the reconstruction and classification algorithms. In this work, we focus on comparing multiple reconstruction and classifier implementations and quantifying the resulting performance. Our analysis shows higher quality reconstructions lead to improved material separability, better classification performance (detection and false alarm rates), and reduces model uncertainty. Through this work, we demonstrate the relationship between improved quality of reconstructions and the separability of threat from non-threat objects in the domain of explosives detection.

Spectral Computed Tomography (CT) is a versatile imaging technique increasingly utilized in industry, medicine, and scientific research. This technique allows us to observe the energy-dependent X-ray attenuation throughout an object by using Photon Counting Detector (PCD) technology. However, a major drawback of Spectral CT is the increase in noise due to a lower photon count per channel, as increasing the number of energy channels without also increasing scan time reduces the photon count per channel. This challenge often complicates quantitative material identification, which is a major application of the technology. In this study, we investigate the use of unsupervised image denoising approaches and demonstrate the applicability of the Noise2Inverse method, an unsupervised denoising method for tomographic imaging. These approaches have the advantage over supervised machine learning methods in that they do not require any additional clean or noisy training data, which can be very difficult to collect in Spectral CT imaging. Our model uses a U-Net paired with a block-based training approach. In particular, we demonstrate that the block-based models can be efficiently trained using small image blocks, each block incorporating spectral information. This training process is performed on images that are reconstructed from subsets of measured Spectral tomography data. The experiments used two simulated Spectral CT phantoms, each with a unique shape and material decomposition. Upon evaluation using the peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) performance metrics, our approach exhibited improvements compared to two alternative approaches: the unsupervised Low2High method previously employed in sparse Spectral CT imaging and a traditional Iterative reconstruction method that imposes a Total Variation (TV) constraint.

As x-ray coherent diffraction-based measurements provide more specific material information, dual modality of x-ray transmission-based and diffraction-based system can provide an improved overall threat detection performance for material discrimination. With limited availability of experimental data collected using real-world objects and materials with a dual-modality x-ray-based machine, in this work, we design and investigate material discrimination performance of two classifiers: support vector machine and neural network using augmented data. The augmented data refers to a combination of experimental data and simulation data, used for training a classifier. The availability of simulation data allows us to sufficiently train a neural network-based classifier which generally require large amount of data.

Annual investments by industry and regulators in screening solutions at airports, seaports, and railway stations is steadily expanding. Over the past two decades, screening technologies have transitioned to complimentary complex detection algorithms designed to detect a wide range of explosive, chemical, and prohibited item threats. There are opportunities to further advance screening solutions through the use of synthetic data. Synthetic data will enable the generation of large and diverse datasets that can be utilized to train and certify enhanced aviation screening detection algorithms. This flexibility allows for a more thorough evaluation of algorithm performance and helps in identifying potential weaknesses or limitations. Synthetic data will assist by creating more accurate and robust screening algorithms that can detect a wider range of threats and anomalies. Synthetic data will help to reduce the need for dangerous, expensive, and time-consuming manual intensive data collection processes. OEMs and 3rd party developers will be able to accelerate the delivery of effective screening capabilities by leveraging readily available and customizable synthetic datasets.

Synthetic data has emerged as a critical piece of the machine-learning based approach to X-ray system design and automatic threat recognition development. Physics-based synthetic data integrates virtual models with a physics-based simulation engine, thereby granting users the capability to produce synthetic measurements based on arbitrary, user-specified input objects and materials. Such inputs can range from geometric phantoms that assist in system design to new threat materials and configurations that expedite ATR training in response to emerging threats. We introduce enhancements to the QSimRT virtual model generation pipeline. This incorporates the rapid creation of virtual models representing passenger luggage, stochastically generated electronics, and user-specified model variability for extensive ensemble production. We have employed these models in training ATRs within the aviation security domain. This presentation will discuss the model generation process, emphasize its pivotal features, and share preliminary results derived from the application of these models in ATR training.

Measuring and validating the quality, visual fidelity, and performance of synthetic image data is an advanced and evolving subject. This study will present an overview of various methods and approaches for measuring the visual quality and fidelity of synthetic image data. Cignal will present an overview of approaches, including established industry standards, such as ASTM E1695 and ANSI N42.45, statistical methods, and neural-network based approaches. Lastly, Cignal will discuss how these approaches may be integrated into a Continuous Integration/ Continuous Delivery (CI/CD) data generation pipeline to monitor and improve data quality and predicted performance for Artificial Intelligence/Machine Learning (AI/ML) use cases.

The Transportation Security Laboratory's Simulant Development Laboratory creates x-ray simulants for explosives and other high-hazard threats. Their patented process employs a MATLAB^®-based optimization program, originally designed for all material types, but recently adapted for putty-based formulas. To simplify putty-based simulant development, researchers explored polymers and dopants to mimic both rheological and x-ray properties. Using an isoprene liquid rubber and powder ingredients, three putty formulations were crafted: one black and two white. The parametric model utilizes xray signatures from different putty simulant formulations, covering the x-ray range of putty-based explosives. Users can select a target point in the x-ray space, which the model correlates with specific ingredients and preloaded putties.

X-ray backscatter (XRBS) is an established technique to conduct through-barrier, non-invasive imaging. Defence Science Technology Laboratory (Dstl) has demonstrated and characterised a novel portable XRBS system tailored for Explosive Ordnance Disposal (EOD), but with broader application across both military and civil contexts. Rapiscan AS&E designed and produced two prototype systems under the direction and funding of US and UK governments. The Portable Backscatter X-ray (PBX2) system is a novel application of the XRBS technique for threat detection, under an innovative form factor which is not currently seen in the commercial market. Dstl has carried out a series of trials covering characterisation of the x-ray performance and identification of potential capability in several threat detection scenarios.

As opposed to transmission imaging, X-ray Phase Contrast Imaging (XPCI) produces images with higher contrast and allows us to distinguish between materials that are weakly attenuating or between materials that possess similar attenuation values. Edge Illumination (EI), a type of XPCI, utilizes spatial variation to uncover information about an object’s phase properties, such as the index of refraction. Instead of spatial variation, we previously proposed an alternative EI method, Spectrally Responsive Edge Illumination (SREI), which relies on energy variation. Prior SREI experimental efforts struggled to meet the necessary component performance requirements, so, as an intermediate step, we are currently focused on developing an energy resolving x-ray refractometer and a related database of materials. In this paper we will share our theory and initial proof of concept experimental results, as well as our next steps.

As part of its mission, the Transportation Security Laboratory (TSL) of the Department of Homeland Security (DHS) Science and Technology Directorate (S&T) develops methods for characterizing materials that are of interest to transportation security and first responders. For emerging technologies, new metrics that meaningfully differentiate materials must be identified and evaluated. Although X-ray diffraction (XRD) is an established technique for identifying solid crystalline materials, the ability to complement the current generation of X-ray-based threat detection using XRD is still being investigated. The TSL has constructed a high-energy X-ray diffraction system to measure a material’s scattering signature, which can vary based on the presence of organic and inorganic materials, solid crystals, and water or other liquids within a sample. Measurements have been performed on a wide range of household items as well as explosives and other threats. The scattering intensity as a function of momentum transfer was examined for each material to identify several potential metrics for distinguishing threats from inert substances.

X-ray phase contrast imaging, which measures the change in phase through an object rather than change in attenuation, excels at distinguishing between similarly attenuating materials when compared to conventional transmission imaging. Most methods of phase contrast imaging, however, are challenging to implement and are limited in application to smaller objects. Previously, we proposed an alternative x-ray phase contrast imaging method based on the asymmetric illumination approach to optical differential phase contrast imaging, which requires only the addition of a single anti-scatter grid. This anti-scatter grid acts as an angular filter that translates asymmetry in the Fourier domain into intensity differences on the detector. Previously, we presented an analytical framework for the method as well as simulation results showing that such a system should be able to obtain phase information. Here, we report on our progress as we continue to advance the experimental implementation and validation of the system. We determine the effect of anti-scatter grid design parameters on the system’s sensitivity to phase change in order to move towards fabrication of a customized anti-scatter grid that utilizes 3D printing capabilities.

The ability to determine the atomic structure or identify the material composition of a sample at high spatial resolution is paramount to a variety of research, imaging, and inspection tasks. We have developed a multi-modality x-ray transmission and x-ray diffraction imaging system that is compact and enables scanning of intact samples. To demonstrate the capabilities of the system, we characterize its spatial and momentum transfer resolution and provide examples of the contrast and utility of the system using a combination of resolution and anthropomorphic phantoms.

We report an experimental demonstration of x-ray Ghost Imaging (GI) to observe the internal structure of a cardamom seed. To the best of our knowledge, this is the first use of GI to see the internal structure of biological samples. GI has been successfully demonstrated at visible and infrared wavelengths as a tool to perform precise imaging. At x-ray energies GI has a further potential to reach this goal with the additional advantage of reducing disruptive radiation doses to biological samples. The experiment was performed using 9.6keV x-rays at the 11-ID Coherent Hard X-ray Scattering (CHX) beamline of the National Synchrotron Light Source II (NSLS-II) facility at Brookhaven National Laboratory (BNL). This imaging technique has application potential in non-destructive examination of sensitive materials in the industrial and commercial sectors.

The Transportation Security Laboratory (TSL) conducts thorough assessments of explosives detection systems (EDSs), encompassing a wide range of explosive materials and hazardous substances. When necessary, inert simulants are employed, but they undergo a stringent verification process to accurately replicate specific properties of threat materials. Whether developed by the TSL or commercially acquired, simulants must undergo verification testing to ensure they mirror the desired threat properties. Historically, these assessments relied on rudimentary metrics like average density and effective atomic number, lacking insight into structural properties possibly being exploited by machine learning detection algorithms. Initial research focused on expanding the verification process by incorporating texture metrics extracted from computed tomography (CT) imagery aimed at deriving features that machine learning detection algorithms might also be utilizing. Two avenues of analysis were devised; first, we calculated 22 metrics through statistical analysis of pixel-based grayscale data, and second, we utilized a convolutional neural network (CNN) to classify images. Both of these methods were subsequently refined and are reported in this work. We augmented the number of metrics for the statistical analysis from 22 to 112, and within the CNN framework we harnessed the flattened array originating from the fully connected layer as a feature map. In both processes the analysis transitioned from a 2-dimensional to a 3-dimensional approach. We assessed the effectiveness of both procedures by testing them on imagery of 50 various materials, such as powders, liquids, putties, and emulsions, using Linear Discriminant Analysis (LDA) to evaluate their ability to distinguish between different materials. Finally, Principal Component Analysis (PCA) loadings were used to define 2-dimensional tolerance intervals for comparisons with loadings from other materials as a way to enhance the current simulant quality control process, ultimately improving the robustness of simulants.

Medical imaging, including the use of chest x-rays, is an important tool for modern healthcare, enabling early and accurate disease diagnosis, facilitating timely interventions to mitigate health issues. By capturing images of critical internal organs like the lungs and heart, x-rays enable doctors to make informed diagnoses and treatment decisions, especially concerning respiratory and cardiac conditions. The importance of early and accurate disease diagnosis, particularly with multiple pathologies, is paramount, as it greatly impacts patient outcomes by enabling timely and specific treatments. Recently, multi-label classification has become increasingly important in medical imaging, since several pathologies can be present within a single x-ray. While traditional convolutional neural networks (CNNs) have played a pivotal role in enhancing the accuracy of x-ray diagnoses, the expanding complexity of multi-label imaging demands more sophisticated methods. Vision Transformers (ViTs) have emerged as a promising approach in medical image classification, showcasing their ability to effectively process x-ray images and identify pathologies within them. While traditional ViTs perform well, they have significant drawbacks. Most ViT models utilize a large number of parameters, often ranging from millions to billions of parameters. Such parameter-intensive designs, while powerful, are computationally heavy. This not only increases the resource requirements, but also raises concerns about their feasibility and scalability in real-world, time-sensitive healthcare settings. We propose a novel Vision Transformer architecture aimed at effectively classifying multi-label x-ray images while significantly enhancing the efficiency of ViT-based multi-label medical image classification methods. By optimizing model architectures and exploring techniques for parameter reduction, we seek to develop more streamlined and resource-efficient approaches without completely sacrificing the efficacy of these methods. Our work endeavors to bridge the gap between cutting-edge technology and practical healthcare applications, promising a more efficient and accessible future for medical image analysis.

Public places, such as airports, factories, public institutions, and organizations, require advanced security measures. X-Ray systems are often used to check passengers’ luggage. Researchers and technologists are constantly working to improve the baggage-checking process by implementing various methods. This study introduces a new method for estimating the effective atomic number. Traditionally, effective atomic number estimation relies on the ratio between high-energy and low-energy images. However, the 3D Effective Atomic Number Estimation model uses high-energy and low-energy image values directly, resolving many issues of the 1D-Ratio model.