Source localization for neutron imaging systems using convolutional neural networks.

The National Ignition Facility (NIF) is scheduled to begin deuterium-tritium (DT) shots possibly in the next several years. One of the important diagnostics in understanding capsule behavior and to guide changes in Hohlraum illumination, capsule design, and geometry will be neutron imaging of both the primary 14 MeV neutrons and the lower-energy downscattered neutrons in the 6-13 MeV range. The neutron imaging system (NIS) described here, which we are currently building for use on NIF, uses a precisely aligned set of apertures near the target to form the neutron images on a segmented scintillator. The images are recorded on a gated, intensified charge coupled device. Although the aperture set may be as close as 20 cm to the target, the imaging camera system will be located at a distance of 28 m from the target. At 28 m the camera system is outside the NIF building. Because of the distance and shielding, the imager will be able to obtain images with little background noise. The imager will be capable of imaging downscattered neutrons from failed capsules with yields Y(n)>10(14) neutrons. The shielding will also permit the NIS to function at neutron yields >10(18), which is in contrast to most other diagnostics that may not work at high neutron yields. The following describes the current NIF NIS design and compares the predicted performance with the NIF specifications that must be satisfied to generate images that can be interpreted to understand results of a particular shot. The current design, including the aperture, scintillator, camera system, and reconstruction methods, is briefly described. System modeling of the existing Omega NIS and comparison with the Omega data that guided the NIF design based on our Omega results is described. We will show NIS model calculations of the expected NIF images based on component evaluations at Omega. We will also compare the calculated NIF input images with those unfolded from the NIS images generated from our NIS numerical modeling code.

Summary form only given. We have installed the National Ignition Facility Neutron Imaging System. The imaging system provides information about the areal density of fuel in the various regions of the capsule implosion. A long line of sight enables imaging both the primary 14 MeV neutrons as well as the down-scattered neutrons with energies in the range of 9-12 MeV. The imager is a pinhole camera where the pinhole is located 325 mm from the target and the imaging plane is located 28 m from the target. The long absorption length of neutrons requires precise alignment of these extended imager components in order to have high detection efficiency and strong background rejection. The imaging plane is a 150 mm square scintillating fiber bundle 50 mm thick. The bundle consists of 250 μm square fibers. The fiber bundle is aligned such that the fibers point at the target. The pointing is accomplished using a retro reflection of a laser alignment device aligned to the imager line of sight. The 200 mm long pinhole is aligned to the imager line of sight using the Opposed Port Alignment System. The imager line of sight was registered to the alignment system to calculate the precise position of the pinhole. The ability of the positioning manipulator to both place and maintain the precise location of the pinhole, were monitored using the positioning imaging system as well as laser alignment device. Here we describe the various aspects of the imaging system alignment to enable the acquisition of neutron images as well as the estimated pointing errors of the first shot imaged.

ACA: Automatic search strategy for radioactive source

SPECT and PET are nuclear tomographic imaging modalities that visualize functional information based on the accumulation of radioactive tracer molecules. However, SPECT and PET lack anatomical information, which has motivated their combination with an anatomical imaging modality such as CT or MRI. This chapter begins with an overview over the fundamental physics of SPECT and PET followed by a presentation of the respective detector technologies, including detection requirements, principles and different detector concepts. The reader is subsequently provided with an introduction into hybrid imaging concepts, before a dedicated section presents the challenges that arise when hybridizing SPECT or PET with MRI, namely, mutual distortions of the different electromagnetic fields in MRI on the nuclear imaging system and vice versa. The chapter closes with an overview about current hybrid imaging systems of both clinical and preclinical kind. Finally, future developments in hybrid SPECT and PET technology are discussed.

Neutron imaging systems measure the spatial distribution of neutron emission from burning inertial confinement fusion (ICF) targets. These systems use a traditional pinhole geometry to project an image of the source onto a two-dimensional scintillator array, and a CCD records the resulting scintillation image. The recent history of ICF neutron images has produced images with qualities that have improved as the fusion neutron yields have increased to nearly 10{sup 14} neutrons. Anticipated future neutron yields in excess of 10{sup 16} at the National Ignition Facility and LMJ have raised the prospect of neuron imaging diagnostics which simultaneously probe several different characteristics of burning fusion targets. The new measurements rely on gated-image recording to select images corresponding to specific bands of neutron energies. Gated images of downscattered neutrons with energies from 5 to 8 MeV can emphasize regions of the target which contain DT fuel which is not burning. At the same time, gated images which select different portions of the 14-MeV spectral peak can produce spatial temperature maps of a burning target. Since the neutron production depends on the DT fuel density and temperature, simultaneous images of temperature and neutron emission can be combined to infer the an image ofmore » the source density using an Abel inversion method that is analogous to the method that has been used in x-ray imaging. Thus, with higher-yield sources, neutron imaging offers the potential to record simultaneously several critical features that characterize the performance of an ICF target: the neutron emission distribution, the temperature and density distributions, and the distribution of nonburning fuel within the target.« less

A deep-learning approach is proposed for simultaneous source localization and seabed classification of underwater signals generated by SUS (Signal, Underwater Sound) MK64 explosives. The charges were deployed in two separate shallow water acoustic experiments on the New England Mudpatch region in 2017 (SBCEX17) and 2022 (SBCEX22) conducted under different oceanographic conditions. The proposed approach uses a multitask learning (MTL) methodology, and convolutional neural networks (CNNs) trained on the spectrogram representation of SUS signals in the 10–199 Hz frequency band where the time-frequency modal dispersion curves exhibit distinct patterns. CNNs were trained on simulated data generated using the normal modes model, ORCA. This synthetic dataset is composed of several sound speed profiles measured in the water column in both experiments and representative seabeds ranging from muddy to sandy sediments. CNNs were tested on the signals measured in the two experiments where the source localization achieved a low mean squared error (0.35–0.8 km), and the seabed classification closely matched existing inversion results in the Mudpatch area. Notably, our method effectively handled oceanographic variations, showcasing the utilization of modal dispersion for training deep learning algorithms. [Work supported by ONR Grant N00014-21-1-2760.]

Passive source localization in shallow water has always been an important and challenging problem. Implementing scientific research, surveying, and monitoring using a short, less than ten meter long, horizontal linear array has received considerable attention in the recent years. The short array can be conveniently placed on autonomous underwater vehicles and deployed for adaptive spatial sampling. However, it is usually difficult to obtain a sufficient spatial gain for localizing long-range sources due to its limited physical size. To address this problem, a localization approach is proposed which is based on matched-field processing of the likelihood of the passive source localization in shallow water, as well as inter-position processing for the improved localization performance and the enhanced stability of the estimation process. The ability of the proposed approach is examined through the two-dimensional synthetic test cases which involves ocean environmental mismatch and position errors of the short array. The presented results illustrate the localization performance for various source locations at different signal- to-noise ratios and demonstrate the build up over time of the positional parameters of the estimated source as the short array moves at a low speed along a straight line at a certain depth.

This study presents a novel approach using Wasserstein Generative Adversarial Networks with Gradient Penalty (WGAN-GP) to generate synthetic electroencephalography (EEG) and electrocardiogram (ECG) waveforms. The synthetic EEG data represent concentration and relaxation mental states, while the synthetic ECG data correspond to normal and abnormal states. By addressing the challenges of limited biophysical data, including privacy concerns and restricted volunteer availability, our model generates realistic synthetic waveforms learned from real data. Combining real and synthetic datasets improved classification accuracy from 92% to 98.45%, highlighting the benefits of dataset augmentation for machine learning performance. The WGAN-GP model achieved 96.84% classification accuracy for synthetic EEG data representing relaxation states and optimal accuracy for concentration states when classified using a fusion of convolutional neural networks (CNNs). A 50% combination of synthetic and real EEG data yielded the highest accuracy of 98.48%. For EEG signals, the real dataset consisted of 60-s recordings across four channels (TP9, AF7, AF8, and TP10) from four individuals, providing approximately 15,000 data points per subject per state. For ECG signals, the dataset contained 1200 real samples, each comprising 140 data points, representing normal and abnormal states. WGAN-GP outperformed a basic generative adversarial network (GAN) in generating reliable synthetic data. For ECG data, a support vector machine (SVM) classifier achieved an accuracy of 98% with real data and 95.8% with synthetic data. Synthetic ECG data improved the random forest (RF) classifier’s accuracy from 97% with real data alone to 98.40% when combined with synthetic data. Statistical significance was assessed using the Wilcoxon signed-rank test, demonstrating the robustness of the WGAN-GP model. Techniques such as discrete wavelet transform, downsampling, and upsampling were employed to enhance data quality. This method shows significant potential in addressing biophysical data scarcity and advancing applications in assistive technologies, human-robot interaction, and mental health monitoring, among other medical applications.

A single photon emission computed tomography (SPECT) system with a co-registered X-y computed tomography (CT) system allows the convergence of functional information and morphologic information. The localization of radiopharmaceuticals on a SPECT can be enhanced by combining the SPECT with an anatomical modality, such as X-ray CT. Gamma-ray imaging for nuclear medicine devices and X-ray imaging systems for diagnostics has recently been developed based on semiconductor detectors, and semiconductor detector materials such as cadmium telluride (CdTe) or cadmium zinc telluride (CZT) are available for both X-ray and gamma-ray systems for small-animal imaging. CdTe or CZT detectors provide strong absorption and high detection efficiency of high energy X-ray and gamma-ray photons because of their large atomic numbers. In this study, a pinhole collimator SPECT system sharing a cadmium telluride (CdTe) detector with a CT was designed. The GEANT4 application for tomographic emission (GATE) v.6.1 was used for the simulation. The pinhole collimator was designed to obtain a high spatial resolution of the SPECT system. The acquisition time for each projection was 40 seconds, and 60 projections were obtained for tomographic image acquisition. The reconstruction was performed using ordered subset expectation maximization (OS-EM) algorithms. The sensitivity and the spatial resolution were measured on the GATE simulation to evaluate the system characteristics. The spatial resolution of the system calculated from the FWHM of Gaussian fitted PSF curve was 0.69 mm, and the sensitivity of the system was measured to be 0.354 cps/kBq by using a Tc-99m point source of 1 MBq for 800 seconds. A phantom study was performed to verify the design of the dual imaging modality system. The system will be built as designed, and it can be applied as a pre-clinical imaging system.

Deep learning (DL) applications of seismic reservoir characterization often require the generation of synthetic data to augment available sparse labeled data. An approach for generating synthetic training data consists of specifying probability distributions modeling prior geologic uncertainty on reservoir properties and forward modeling the seismic data. A prior falsification approach is critical to establish the consistency of the synthetic training data distribution with real seismic data. With the help of a real case study of facies classification with convolutional neural networks (CNNs) from an offshore deltaic reservoir, we have highlighted several practical nuances associated with training DL models on synthetic seismic data. We highlight the issue of overfitting of CNNs to the synthetic training data distribution and propose regularization strategies to address it. We demonstrate the efficacy of our proposed strategies by training the CNN on synthetic data and making robust predictions with real 3D partial stack seismic data.

The ample size of time-lapse data often requires significant event detection and source location efforts, especially in areas such as shale gas exploration regions where a large number of microseismic events are often recorded. In many cases, real-time monitoring and location of these events are essential to production decisions. Conventional methods face considerable drawbacks. For example, traveltime-based methods require traveltime picking of often noisy data, whereas migration and waveform inversion methods require expensive wavefield solutions and event detection. Both tasks require some human intervention, and this becomes a big problem when too many sources need to be located, which is common in microseismic monitoring. Machine learning has recently been used to identify microseismic events or locate their sources once they are identified and picked. We have used a novel artificial neural network framework to directly map seismic data, without any event picking or detection, to their potential source locations. We train two convolutional neural networks (CNNs) on labeled synthetic acoustic data containing simulated microseismic events to fulfill such requirements. One CNN, which has a global average pooling layer to reduce the computational cost while maintaining high-performance levels, aims to classify the number of events in the data. The other network predicts the source locations and other source features such as the source peak frequencies and amplitudes. To reduce the size of the input data to the network, we correlate the recorded traces with a central reference trace to allow the network to focus on the curvature of the input data near the zero-lag region. We train the networks to handle single-, multi-, and no-event segments extracted from the data. Tests on a simple vertical varying model and a more realistic Otway field model demonstrate the approach’s versatility and potential.

This study assesses the effectiveness of deep learning models, specifically Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM), in assessing synthetic grip strength data for stress ball and pencil grips. The aim is to determine the optimal model for grip strength classification and tackle issues such as temporal data and class imbalance. Synthetic data was created, preprocessed, and utilized to train the models on Google Colab employing TensorFlow and Keras. Hyperparameters, such as CNN kernel dimensions and RNN/LSTM sequence lengths, were refined. Performance was evaluated by accuracy, precision, F1 scores, and loss metrics. The RNN model surpassed both CNN and LSTM, with an accuracy of 94.99% and a precision of 0.56, particularly excelling in the classification of “Normal” grip strength instances. CNN and LSTM had difficulties with time dependencies and class imbalance, leading to reduced accuracy and increased loss. Although RNN has proven to be the most effective approach, subsequent research should concentrate on mitigating class imbalance and investigating hybrid models to improve performance.

This paper presents a learning-based method for source localization in the presence of directional interference under reverberant and noisy conditions. The proposed method operates on the spherical harmonic decomposition of the spherical microphone array recordings to yield spherical harmonics coefficients as the features. An attention mechanism is incorporated through a binary mask that filters out the dominant undesired source components from the features before training. A convolutional neural network is trained to map the phase and magnitude of the filtered coefficients with the location class. Hence, the objective is to develop the binary mask followed by source localization.

To solve the problem that phase information cannot be used effectively in underwater acoustic localization, a deep learning network based on complex convolutional neural network is proposed in this paper. The complex convolution layer is used to effectively utilize the phase features favorable to the source localization problem, and the phase information is used to improve the feature extraction performance. Through simulation, the performance of the network for source localization in deep-sea direct arrival region under different SNR conditions is analyzed. The results show that the complex convolutional network proposed in this paper can locate the sound source with less computation and has better performance under the condition of low SNR.

Brain tumors are formed due to the abnormal growth of the cells that get multiplied and become in uncontrollable perspective. Tumors can damage brain cells by pressing down the skull, which consociate begins in, which negatively affects human health. In advanced stages, a brain tumor is a more dangerous infection that cannot be relieved. Brain tumor detection and early prevention are necessary in today's world. An extreme learning machine (ELM) is a widely adopted algorithm in machine learning. It is proposed to use classification models in brain tumor imaging. This classification is based on the techniques implemented: Convolutional Neural Networks(CNN) and Generative Adversarial Networks(GAN). CNN efficiently solves the convex optimization problem and is faster, requiring less human effort. The algorithmic architecture of GAN uses two neural networks, pocking one against the other. These networks are implemented in various fields to classify brain Tumor images. The present study mainly aims to introduce a new proposed classification system of preschool children brain imaging with Hybrid Convolutional Neural Networks and with the techniques of GAN. The proposed technique is compared with the existing hybrid-CNN and hybrid-GAN techniques. The outcomes are encouraging because the loss is deduced, and the accuracy facet increases. The proposed system achieved a training accuracy of 97.8% and a validation accuracy of 89%. The outcomes of the studies show that ELM in the platform of GAN for preschool children brain imaging classification has achieved higher predictive performance than the traditional classification mechanisms in increasingly complex scenarios. Time elapsed for training brain images samples finds inference value for training samples and time elapsed value increased by 28.9855%. Approximation ratio for cost based on probability, finding Approximation ratio for low probability range is increased by 88.1%. The combination of CNN, GAN and hybrid-CNN, hybrid-GAN, and hybrid CNN + GAN, compared with the proposed hybrid system, increased Detection latency for low range learning rate by 3.31%.