Real-time Tomography Research Articles

Just-in-time deep learning for real-time X-ray computed tomography

Real-time X-ray tomography pipelines, such as implemented by RECAST3D, compute and visualize tomographic reconstructions in milliseconds, and enable the observation of dynamic experiments in synchrotron beamlines and laboratory scanners. For extending real-time reconstruction by image processing and analysis components, Deep Neural Networks (DNNs) are a promising technology, due to their strong performance and much faster run-times compared to conventional algorithms. DNNs may prevent experiment repetition by simplifying real-time steering and optimization of the ongoing experiment. The main challenge of integrating DNNs into real-time tomography pipelines, however, is that they need to learn their task from representative data before the start of the experiment. In scientific environments, such training data may not exist, and other uncertain and variable factors, such as the set-up configuration, reconstruction parameters, or user interaction, cannot easily be anticipated beforehand, either. To overcome these problems, we developed just-in-time learning, an online DNN training strategy that takes advantage of the spatio-temporal continuity of consecutive reconstructions in the tomographic pipeline. This allows training and deploying comparatively small DNNs during the experiment. We provide software implementations, and study the feasibility and challenges of the approach by training the self-supervised Noise2Inverse denoising task with X-ray data replayed from real-world dynamic experiments.

Inclusion of environmental stochasticity in navy system performance modeling

The performance of all acoustic navy systems is criticallly dependent upon the environmental conditions for propagation, primarily the ocean sound speed structure, the seafloor topography, the sediment type and the sea surface characteristics. Advances in computational power and improvements in hydrodynamic modeling have led to the opportunity of including ocean varaibiltiy and uncertainty into the acoustic propagation portion of system performance and, therefore, through the application of straight forward statistics into the performance prediction product. The first application of stochastic processing is in the computation of ambient noise (the shipping and noise sounsdcape) for comparison between modeled noise and measurements from long-term buoy deployments. Uncertainty in ship source level, position, wind speed, and environmental variabiility is key drivers in the model-data comparison. A second example of inclusion of environmental stochasticity in the passive anti-submarine warfare scenario of a passive surveillance barrier. A primary challenge here is in the communication of the uncertainty to a community that is used to thinking in simple performance metrics, such as range of the day. The last example is the use of an ensemble ocean forecast in the computation of the spread in eigenray arrivals for a real-time tomography system utilizing an active ASW sonobuoy field.

Detecting illegal wildlife trafficking via real time tomography 3D X-ray imaging and automated algorithms

Wildlife trafficking is a global problem involving the deliberate and illegal transport of wildlife across international borders. Animals are either removed directly from their natural environment or bred specifically to fuel demand driven by activities such as the illegal pet trade or for purported medicinal reasons. In Australia, wildlife trafficking poses a serious environmental and biosecurity risk through the removal of native species and the introduction of exotic invasive wildlife. This has the potential to impact the natural ecosystem and Australia’s multibillion-dollar agricultural industry. To help detect and restrict this activity, innovative technologies such as 3D X-ray CT technology using Real Time Tomography has been trialed to create wildlife detection algorithms for deployment across Australian mail/traveller luggage pathways. Known species of trafficked Australian wildlife and additional model species of exotics were scanned to create an image reference library for algorithm detection. A total of 294 scans from 13 species of lizards, birds and fish were used to develop initial wildlife algorithms with a detection rate of 82% with a false alarm rate at 1.6%. In combination with human and biosecurity dog detection, this innovative technology is a promising complementary platform for wildlife detection at Australian international borders, with potential worldwide applications.

Real-time 3D analysis during electron tomography using tomviz

The demand for high-throughput electron tomography is rapidly increasing in biological and material sciences. However, this 3D imaging technique is computationally bottlenecked by alignment and reconstruction which runs from hours to days. We demonstrate real-time tomography with dynamic 3D tomographic visualization to enable rapid interpretation of specimen structure immediately as data is collected on an electron microscope. Using geometrically complex chiral nanoparticles, we show volumetric interpretation can begin in less than 10 minutes and a high-quality tomogram is available within 30 minutes. Real-time tomography is integrated into tomviz, an open-source and cross-platform 3D data analysis tool that contains intuitive graphical user interfaces (GUI), to enable any scientist to characterize biological and material structure in 3D.

Real-time tomography of the human brain.

Real-time tomography of the human brain.

Investigating the physics of disruptions with real-time tomography at JET

As JET is developing and testing operational scenarios for higher fusion performance, an increase in pulse disruptivity is being observed. On a deeper analysis, we find that several radiative phenomena play an active role in determining the outcome of the pulse. The analysis is enabled by the use of real-time tomography based on the bolometer diagnostic. Even though plasma tomography is an inverse problem, we use machine learning to train a forward model that provides the radiation profile directly, based on a single matrix multiplication step. This model is used to investigate radiative phenomena including sawtooth crashes, ELMs and MARFE, and their relationship to the radiated power in different regions of interest. In particular, we use real-time tomography to monitor the core region, and to throw an alarm whenever core radiation exceeds a certain threshold. Our results suggest that this measure alone can anticipate a significant fraction of disruptions in the JET baseline scenario.

Monitoring the plasma radiation profile with real-time bolometer tomography at JET

The use of real-time tomography at JET opens up new possibilities for monitoring the plasma radiation profile and for taking preventive or mitigating actions against impending disruptions. By monitoring the radiated power in different plasma regions, such as core, edge and divertor, it is possible to set up multiple alarms for the radiative phenomena that usually precede major disruptions. The approach is based on the signals provided by the bolometer diagnostic. Reconstructing the plasma radiation profile from these signals is a computationally intensive task, which is typically performed during post-pulse analysis. To reconstruct the radiation profile in real-time, we use machine learning to train a surrogate model that performs matrix multiplication over the bolometer signals. The model is trained on a large number of sample reconstructions, and is able to compute the plasma radiation profile within a few milliseconds in real-time. The implementation has been further optimized by computing the radiated power only in the regions of interest. Experimental results show that, during uncontrolled termination, there is an impurity accumulation at the plasma core, which eventually leads to a disruption. A threshold-based alarm on core radiation, among other options, is able to anticipate a significant fraction of such disruptions.

Development of injectable thermosensitive polypeptide hydrogel as facile radioisotope and radiosensitizer hotspot for synergistic brachytherapy

Brachytherapy is considered to be an unparalleled form of conformal radiation therapy, which involves the delivery of radiation directly to tumor lesions or the postoperative cavity. With the development of specific applicators, the exploitation of in situ drug-delivery platform introduces opportunities for the synchronous administration of radiosensitizers. In this study, an iodine-131 (I131)-labeled injectable thermosensitive methoxy poly(ethylene glycol)-b-poly(tyrosine) hydrogel (denoted as PETyr-I131) was developed via a facile method. The radioactive source of I131 was immobilized at the subcutaneous injection site and monitored via single-photon emission computed tomography in real time, and hematological and histopathological analyses revealed no obvious side effects. Additionally, the SmacN7 peptide conjugated with cell membrane-permeable oligosarginine (denoted as SmacN7-R9) was used to enhance the radiosensitivity of cancer cells, as confirmed by the results of reactive oxygen species detection, DNA damage assay, cell apoptosis assay, and clonogenic evaluation. Importantly, a synergistic brachytherapy treatment effect on tumor-bearing nude mice was achieved. The proposed thermosensitive supramolecular hydrogel platform, which conformally immobilizes radionuclides and delivers radiosensitizers by virtue of its proximity to the site of the primary tumor or the postoperative cavity, has great potential for achieving synergistic treatment outcomes with reduced radiation-related side effects. Statement of SignificanceIn this work, a kind of radioiodinated thermosensitive supramolecular hydrogel was developed, which was facilely used as the radioactive source for brachytherapy. Meanwhile, SmacN7-R9 peptide was combined as a model radiosensitizer to facilitate the activation of tumor cell apoptosis pathways and promotion of radiation-induced cytotoxicity. Synergistic brachytherapy outcomes were achieved from the in vitro and in vivo evaluations. Therefore, from the practical standpoint, this thermosensitive supramolecular hydrogel platform holds great potential for the 3D-conformally immobilizing radionuclide and delivering radiosensitizer by virtue of its proximity to the site of primary tumor lesions or postoperative cavity, resulting in synergetic treatment outcomes with reduced radiation associated side effects.

A Data Consistent Variational Segmentation Approach Suitable for Real-time Tomography

Computed Tomography (CT) is an imaging technique that allows to reconstruct volumetric information of the analyzed objects from their projections. The most popular reconstruction technique is the Filtered Back Projection (FBP). It has the advantage of being the fastest technique available, but also the disadvantage to require a high number of projections to retrieve good quality reconstructions. In this article we propose a segmentation method for tomographic volumes composed of few materials. Our method combines existing high-quality variational segmentation frameworks with the data consistency approach used in tomography and discrete tomography. We show that our algorithm performs well under high noise level and with moderately low number of projections, and that the data consistency significantly improves the segmentation, at the cost of only one FBP reconstruction and forward projection.

A Real-Time Hard X-Ray Tomographic System During Lower Hybrid Current Drive on HL-2A Tokamak

A set of real-time tomographic systems for tokamak plasma hard X-ray (HXR) diagnosis was designed, which is dedicated to real-time investigation of X-ray energy and its density profile that are produced by fast electron bremsstrahlung emission between the energy span of 20 to 200 keV under the lower hybrid (LH)–driven mode of HL-2A device. By utilizing an integrated detector array of lutetium yttrium oxyorthosilicate scintillators and silicon photomultipliers, and in conjunction with high-speed main electronics and appropriate digital processing algorithms, this tomographic system has achieved real-time energy surveillance of HXR that are radiated from 12 different chords on the poloidal plane of the HL-2A plasma region. The temporal and spatial resolutions of this system are 10 ms and 2 cm, respectively. The real-time tomography of the investigation results can clearly display the spatial distribution and its evolution over time of HXR energy. Onsite HL-2A experiments have proven that this real-time tomographic system is qualified as a reliable platform in the research field of plasma state and high-energy particles energy during LH current mode.

Distributed tomography with adaptive mesh refinement in sensor networks

Existing seismic instrumentation systems do not yet have the capability to recover the physical dynamics with sufficient resolution in real time. Currently, seismologists use centralised tomography inversion algorithm, which requires manual data gathering from each station and months to generate tomography. To address these issues a distributed approach is required which can avoid data collection from large number of sensors and perform in-network imaging to real-time tomography. In this paper, we present a distributed adaptive mesh refinement AMR solution to invert seismic tomography over large dense network, which avoids centralised computation and expensive data collection. Our approach first discretises the data and filters them using adaptive mesh to make it well-conditioned. The system is implemented and evaluated using a CORE emulator and we show that the filtered well-conditioned system has lower dimension and improved convergence rate than the original system, thereby decreasing the communication overhead over the network.

Image Enhancement Using Active Contour Filtering Methods for a Custom-Developed Real-Time Photoacoustic Tomography System to Accurately Delineate a Target

Image Enhancement Using Active Contour Filtering Methods for a Custom-Developed Real-Time Photoacoustic Tomography System to Accurately Delineate a Target

MO-DE-209-01: Primer On Tomosynthesis

Digital Breast Tomosynthesis (DBT) is rapidly replacing mammography as the standard of care in breast cancer screening and diagnosis. DBT is a form of computed tomography, in which a limited set of projection images are acquired over a small angular range and reconstructed into tomographic data. The angular range varies from 15° to 50° and the number of projections varies between 9 and 25 projections, as determined by the equipment manufacturer. It is equally valid to treat DBT as the digital analog of classical tomography – that is, linear tomography. In fact, the name “tomosynthesis” stands for “synthetic tomography.” DBT shares many common features with classical tomography, including the radiographic appearance, dose, and image quality considerations. As such, both the science and practical physics of DBT systems is a hybrid between computed tomography and classical tomographic methods. In this lecture, we will explore the continuum from radiography to computed tomography to illustrate the characteristics of DBT. This lecture will consist of four presentations that will provide a complete overview of DBT, including a review of the fundamentals of DBT acquisition, a discussion of DBT reconstruction methods, an overview of dosimetry for DBT systems, and summary of the underlying image theory of DBT thereby relating image quality and dose. Learning Objectives: 1. To understand the fundamental principles behind tomosynthesis image acquisition. 2. To understand the fundamentals of tomosynthesis image reconstruction. 3. To learn the determinants of image quality and dose in DBT, including measurement techniques. 4. To learn the image theory underlying tomosynthesis, and the relationship between dose and image quality. ADM is a consultant to, and holds stock in, Real Time Tomography, LLC. ADM receives research support from Hologic Inc., Analogic Inc., and Barco NV.; ADM is a member of the Scientific Advisory Board for Gamma Medica Inc.; A. Maidment, Research Support, Hologic, Inc.; Research Support, Barco, Inc.; Scientific Advisory Board, Gamma Medica, Inc.; Scientific Advisory Board, Real-Time Tomography, LLC.; Shareholder, Real-Time Tomography, LLC; J. Mainprize, Our lab has a research agreement with GE Healthcare on various topics in digital mammography and digital tomosynthesis; W. Zhao, Research grant from Siemens Health Care

MO-DE-209-02: Tomosynthesis Reconstruction Methods

Digital Breast Tomosynthesis (DBT) is rapidly replacing mammography as the standard of care in breast cancer screening and diagnosis. DBT is a form of computed tomography, in which a limited set of projection images are acquired over a small angular range and reconstructed into tomographic data. The angular range varies from 15° to 50° and the number of projections varies between 9 and 25 projections, as determined by the equipment manufacturer. It is equally valid to treat DBT as the digital analog of classical tomography – that is, linear tomography. In fact, the name “tomosynthesis” stands for “synthetic tomography.” DBT shares many common features with classical tomography, including the radiographic appearance, dose, and image quality considerations. As such, both the science and practical physics of DBT systems is a hybrid between computed tomography and classical tomographic methods. In this lecture, we will explore the continuum from radiography to computed tomography to illustrate the characteristics of DBT. This lecture will consist of four presentations that will provide a complete overview of DBT, including a review of the fundamentals of DBT acquisition, a discussion of DBT reconstruction methods, an overview of dosimetry for DBT systems, and summary of the underlying image theory of DBT thereby relating image quality and dose. Learning Objectives: 1. To understand the fundamental principles behind tomosynthesis image acquisition. 2. To understand the fundamentals of tomosynthesis image reconstruction. 3. To learn the determinants of image quality and dose in DBT, including measurement techniques. 4. To learn the image theory underlying tomosynthesis, and the relationship between dose and image quality. ADM is a consultant to, and holds stock in, Real Time Tomography, LLC. ADM receives research support from Hologic Inc., Analogic Inc., and Barco NV.; ADM is a member of the Scientific Advisory Board for Gamma Medica Inc.; A. Maidment, Research Support, Hologic, Inc.; Research Support, Barco, Inc.; Scientific Advisory Board, Gamma Medica, Inc.; Scientific Advisory Board, Real-Time Tomography, LLC.; Shareholder, Real-Time Tomography, LLC; J. Mainprize, Our lab has a research agreement with GE Healthcare on various topics in digital mammography and digital tomosynthesis; W. Zhao, Research grant from Siemens Health Care

MO-DE-209-00: DBT Physics Basic to Advanced

Digital Breast Tomosynthesis (DBT) is rapidly replacing mammography as the standard of care in breast cancer screening and diagnosis. DBT is a form of computed tomography, in which a limited set of projection images are acquired over a small angular range and reconstructed into tomographic data. The angular range varies from 15° to 50° and the number of projections varies between 9 and 25 projections, as determined by the equipment manufacturer. It is equally valid to treat DBT as the digital analog of classical tomography – that is, linear tomography. In fact, the name “tomosynthesis” stands for “synthetic tomography.” DBT shares many common features with classical tomography, including the radiographic appearance, dose, and image quality considerations. As such, both the science and practical physics of DBT systems is a hybrid between computed tomography and classical tomographic methods. In this lecture, we will explore the continuum from radiography to computed tomography to illustrate the characteristics of DBT. This lecture will consist of four presentations that will provide a complete overview of DBT, including a review of the fundamentals of DBT acquisition, a discussion of DBT reconstruction methods, an overview of dosimetry for DBT systems, and summary of the underlying image theory of DBT thereby relating image quality and dose. Learning Objectives: 1.To understand the fundamental principles behind tomosynthesis image acquisition. 2.To understand the fundamentals of tomosynthesis image reconstruction. 3.To learn the determinants of image quality and dose in DBT, including measurement techniques. 4.To learn the image theory underlying tomosynthesis, and the relationship between dose and image quality. ADM is a consultant to, and holds stock in, Real Time Tomography, LLC. ADM receives research support from Hologic Inc., Analogic Inc., and Barco NV.; ADM is a member of the Scientific Advisory Board for Gamma Medica Inc.; A. Maidment, Research Support, Hologic, Inc.; Research Support, Barco, Inc.; Scientific Advisory Board, Gamma Medica, Inc.; Scientific Advisory Board, Real-Time Tomography, LLC.; Shareholder, Real-Time Tomography, LLC; J. Mainprize, Our lab has a research agreement with GE Healthcare on various topics in digital mammography and digital tomosynthesis; W. Zhao, Research grant from Siemens Health Care

MO-DE-209-03: Assessing Image Quality

Digital Breast Tomosynthesis (DBT) is rapidly replacing mammography as the standard of care in breast cancer screening and diagnosis. DBT is a form of computed tomography, in which a limited set of projection images are acquired over a small angular range and reconstructed into tomographic data. The angular range varies from 15° to 50° and the number of projections varies between 9 and 25 projections, as determined by the equipment manufacturer. It is equally valid to treat DBT as the digital analog of classical tomography – that is, linear tomography. In fact, the name “tomosynthesis” stands for “synthetic tomography.” DBT shares many common features with classical tomography, including the radiographic appearance, dose, and image quality considerations. As such, both the science and practical physics of DBT systems is a hybrid between computed tomography and classical tomographic methods. In this lecture, we will explore the continuum from radiography to computed tomography to illustrate the characteristics of DBT. This lecture will consist of four presentations that will provide a complete overview of DBT, including a review of the fundamentals of DBT acquisition, a discussion of DBT reconstruction methods, an overview of dosimetry for DBT systems, and summary of the underlying image theory of DBT thereby relating image quality and dose. Learning Objectives: 1.To understand the fundamental principles behind tomosynthesis image acquisition. 2.To understand the fundamentals of tomosynthesis image reconstruction. 3.To learn the determinants of image quality and dose in DBT, including measurement techniques. 4.To learn the image theory underlying tomosynthesis, and the relationship between dose and image quality. ADM is a consultant to, and holds stock in, Real Time Tomography, LLC. ADM receives research support from Hologic Inc., Analogic Inc., and Barco NV.; ADM is a member of the Scientific Advisory Board for Gamma Medica Inc.; A. Maidment, Research Support, Hologic, Inc.; Research Support, Barco, Inc.; Scientific Advisory Board, Gamma Medica, Inc.; Scientific Advisory Board, Real-Time Tomography, LLC.; Shareholder, Real-Time Tomography, LLC; J. Mainprize, Our lab has a research agreement with GE Healthcare on various topics in digital mammography and digital tomosynthesis; W. Zhao, Research grant from Siemens Health Care

MO-DE-209-04: Radiation Dosimetry in Breast Tomosynthesis

Digital Breast Tomosynthesis (DBT) is rapidly replacing mammography as the standard of care in breast cancer screening and diagnosis. DBT is a form of computed tomography, in which a limited set of projection images are acquired over a small angular range and reconstructed into tomographic data. The angular range varies from 15° to 50° and the number of projections varies between 9 and 25 projections, as determined by the equipment manufacturer. It is equally valid to treat DBT as the digital analog of classical tomography – that is, linear tomography. In fact, the name “tomosynthesis” stands for “synthetic tomography.” DBT shares many common features with classical tomography, including the radiographic appearance, dose, and image quality considerations. As such, both the science and practical physics of DBT systems is a hybrid between computed tomography and classical tomographic methods. In this lecture, we will explore the continuum from radiography to computed tomography to illustrate the characteristics of DBT. This lecture will consist of four presentations that will provide a complete overview of DBT, including a review of the fundamentals of DBT acquisition, a discussion of DBT reconstruction methods, an overview of dosimetry for DBT systems, and summary of the underlying image theory of DBT thereby relating image quality and dose. Learning Objectives: 1. To understand the fundamental principles behind tomosynthesis image acquisition. 2. To understand the fundamentals of tomosynthesis image reconstruction. 3. To learn the determinants of image quality and dose in DBT, including measurement techniques. 4. To learn the image theory underlying tomosynthesis, and the relationship between dose and image quality. ADM is a consultant to, and holds stock in, Real Time Tomography, LLC. ADM receives research support from Hologic Inc., Analogic Inc., and Barco NV.; ADM is a member of the Scientific Advisory Board for Gamma Medica Inc.; A. Maidment, Research Support, Hologic, Inc.; Research Support, Barco, Inc.; Scientific Advisory Board, Gamma Medica, Inc.; Scientific Advisory Board, Real-Time Tomography, LLC.; Shareholder, Real-Time Tomography, LLC; J. Mainprize, Our lab has a research agreement with GE Healthcare on various topics in digital mammography and digital tomosynthesis; W. Zhao, Research grant from Siemens Health Care

Minimum Fisher Tikhonov Regularization Adapted to Real-Time Tomography

Tomography inversion has been used routinely for processing outputs of plasma radiation diagnostics. Various tomographic algorithms have been developed, with those based on Tikhonov regularization being among the fastest while still providing reliable results. This paper presents a further speed optimization of the minimum Fisher Tikhonov regularization algorithm based on reducing iteration cycles used during the calculation. Ten to twentyfold speed gain is achieved compared to the original implementation. Robustness of the new method is demonstrated using both artificially generated data sets and real data from the soft X-ray diagnostics at the COMPASS tokamak. The advantage gained by the optimization is investigated in particular with respect to the possibility of real-time control of the plasma position; the option of impurity control is also discussed.

Towards in-process x-ray CT for dimensional metrology

X-ray computed tomography (CT) offers significant potential as a metrological tool, given the wealth of internal and external data that can be captured, much of which is inaccessible to conventional optical and tactile coordinate measurement machines (CMM). Typical lab-based CT can take upwards of 30 min to produce a 3D model of an object, making it unsuitable for volume production inspection applications. Recently a new generation of real time tomography (RTT) x-ray CT has been developed for airport baggage inspections, utilising novel electronically switched x-ray sources instead of a rotating gantry. This enables bags to be scanned in a few seconds and 3D volume images produced in almost real time for qualitative assessment to identify potential threats. Such systems are able to scan objects as large as 600 mm in diameter at 500 mm s−1. The current voxel size of such a system is approximately 1 mm—much larger than lab-based CT, but with significantly faster scan times is an attractive prospect to explore. This paper will examine the potential of such systems for real time metrological inspection of additively manufactured parts. The measurement accuracy of the Rapiscan RTT110, an RTT airport baggage scanner, is evaluated by comparison to measurements from a metrologically confirmed CMM and those achieved by conventional lab-CT. It was found to produce an average absolute error of 0.18 mm that may already have some applications in the manufacturing line. While this is expectedly a greater error than lab-based CT, a number of adjustments are suggested that could improve resolution, making the technology viable for a broader range of in-line quality inspection applications, including cast and additively manufactured parts.

A new systolic multiprocessor architecture for real-time soft tomography algorithms

A fixed-point reconfigurable systolic VLSI hardware architecture for real-time ECT.Up to 8.475?Kframes/sec a throughput for the Landweber algorithm could be achieved.The suggested architecture satisfies the real-time requirements of several applications. In this paper, a new systolic multiprocessor architecture for soft tomography algorithms that explores the intrinsic parallelisms and hardware resources which are available in recent Field Programmable Gate Arrays architectures is presented. The soft tomography algorithms such as Electrical Capacitance Tomography (ECT), Magnetic Inductance Tomography (MIT), and Electrical Impedance Tomography (EIT), while they use different sensors and data acquisition modules, they feature common computation requirements which consist of intensive matrix multiplications and fast/frequent memory accesses. Using the variable bit-width and fixed-point multipliers array available in the DSP blocks, which cooperatively perform the partial matrix product with associated Arithmetic and Logic Units (ALU), and distributed memory available in Stratix V FPGA, a dedicated scalable architecture is suggested to host the Landweber algorithm. The experimental results indicate that 16,949 frames of (32?×?32 pixels) can be reconstructed in one second if each element of the matrix is attributed to 18?bits and using a clock frequency of 400?MHz. This is more than enough in most process imaging applications. In addition, the accuracy of the image reconstruction using 18?bits/operand is found to be acceptable since it exceeds 86%. More accuracy can be achieved up to 99% if 36 bits/operand are used which leads to an image reconstruction throughput of 1272 frames /s (for image size 32?×?32).