Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “medical image processing”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

High-Throughput Computing: Case Study of Medical Image Processing Applications

HPC is designed for large-scale simulations using monolithic codes of tightly coupled processes highly optimized to deliver decreased time to solution. Medical image processing is not a traditional field of HPC. Similar to AI applications, medical image processing parses large datasets, typically multiple times, to support a variety of studies for classification, diagnosis or monitoring purposes. The convergence of AI, HPC and Big Data encouraged more fields using image processing to transition to HPC. However, not all applications benefit from the same optimizations. In this paper we focus on high throughput medical image processing applications that analyze a huge dataset of small MRI images and that require HPC systems to decrease the time of parsing the entire dataset and not individual MRIs. We show in this research the performance of running SLANT, an image processing application for a whole brain segmentation, on large-scale systems and highlight performance limitations. We present optimizations prioritizing throughput that exhibit a 3.5x speed-up on the Summit Supercomputer that can be used as a baseline for building a high-throughput execution framework for other HPC systems.

Predescu, Maria↗
DOE OSTI2024

Quapprox: A Framework for Benchmarking the Approximability of Variational Quantum Circuit

Most of the existing quantum neural network models, such as variational quantum circuits (VQCs), are limited in their ability to explore the non-linear relationships in input data. This gradually becomes the main obstacle for it to tackle realistic applications, such as natural language processing, medical image processing, and wireless communications. Recently, there have emerged research efforts that enable VQCs to perform non-linear operations. However, it is still unclear on the approximability of a given VQC (i.e., the order of non-linearity that can be handled by a specified design). In response to this issue, we developed an automated tool designed to benchmark the approximation of a given VQC. The proposed tool will generate a set of synthetic datasets with different orders of non-linearity and train the given VQC on these datasets to estimate their approximability. Our experiments benchmark VQCs with different designs, where we know their theoretic approximability. We then show that the proposed tool can precisely estimate the approximability, which is consistent with the theoretic value, indicating that the proposed tool can be used for benchmarking the approximability of a given quantum circuit for learning tasks.

artificial intelligence↗
DOE OSTI2019

Iterative Joint Ptychography-Tomography with Total Variation Regularization

In order to determine the 3D structure of a thick sample, researchers have recently combined ptychography (for high resolution) and tomography (for 3D imaging) in a single experiment. 2-step methods are usually adopted for reconstruction, where the ptychography and tomography problems are often solved independently. In this paper, we provide a novel model and ADMM-based algorithm to jointly solve the ptychography-tomography problem iteratively, also employing total variation regularization. The proposed method permits large scan stepsizes for the ptychography experiment, requiring less measurements and being more robust to noise with respect to other strategies, while achieving higher reconstruction quality results.

2-step methods↗
DOE OSTI2018

Collaborative and Reproducible Research: Goals, Challenges, and Strategies

Combining imaging biomarkers with genomic and clinical phenotype data is the foundation of precision medicine research efforts. Yet, biomedical imaging research requires unique infrastructure compared with principally text-driven clinical electronic medical record (EMR) data. The issues are related to the binary nature of the file format and transport mechanism for medical images as well as the post-processing image segmentation and registration needed to combine anatomical and physiological imaging data sources. The SiiM Machine Learning Committee was formed to analyze the gaps and challenges surrounding research into machine learning in medical imaging and to find ways to mitigate these issues. At the 2017 annual meeting, a whiteboard session was held to rank the most pressing issues and develop strategies to meet them. The results, and further reflections, are summarized in this paper.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2020

Octave bandwidth photonic fishnet-achromatic-metalens

Planar structured interfaces, also known as metasurfaces, are continuously attracting interest owing to their ability to manipulate fundamental attributes of light, including angular momentum, phase, or polarization. However, chromatic aberration, limiting broadband operation, has remained a challenge for metasurfaces-based optical components and imagers. The limitation stems from the intrinsic dispersion of existing materials and design principles. Here we report and experimentally demonstrate polarization-independent fishnet-achromatic-metalenses with measured average efficiencies over 70% in the continuous band from the visible (640 nm) to the infrared (1200 nm). Results of the scalable platform are enabling for applications requiring broad bandwidth and high efficiency including energy harvesting, virtual reality and information processing devices, or medical imaging.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Terahertz time-domain spectroscopy imaging of pancreatic ductal adenocarcinoma tissues

Pancreatic ductal adenocarcinoma (PDAC) ranks among the malignancies with the highest fatality and morbidity rates. This is predominantly attributable to an absence of understanding the intricate and diverse microenvironment of the tumor. We use terahertz time-domain spectroscopy (THz-TDS) imaging in transmission geometry to probe ex-vivo the heterogenous microenvironment of the genetically modified murine PDAC tissue that closely resembles the PDAC heterogeneity in human malignancy. We introduced a maximum a-posteriori probability estimation algorithm to objectively the tumor’s heterogenous microenvironment using the average values of refractive index and absorption coefficient within the useable terahertz bandwidth as imaging markers. Furthermore, direct comparison of stained histopathologic images and the refractive index and the absorption coefficient high-resolution, two dimensional maps of the same PDAC samples confirms the high potential of the THz-TDS method for tumor tissue characterization.

absorption↗
DOE OSTI2024

Machine learning based reconstruction of intracardiac electrical behavior based on electrocardiograms

A computer-based system and process are disclosed for reconstructing the internal electrical behavior of a patient's heart based partly or wholly on the patient's electrocardiogram (ECG). The output of the process may include, for example, a cardiac activation map, and/or a representation of transmembrane potentials over time. The process advantageously does not require any medical imaging of the patient, and does not require any special medical equipment. For example, the patient's activation map and transmembrane potentials may be reconstructed based solely on a preexisting or newly-obtained 12-lead cardiac ECG of the patient. The process makes use of a machine learning model, such as a neural network based model, trained with actual and/or simulated ECGs and intracardiac electrical data (typically transmembrane potentials) of many thousands of patients. Because an insufficient quantity of such data exists for actual patients, model training may be performed using ECGs and intracardiac electrical data obtained through computer simulations.

Blake, Robert↗
DOE OSTI2018

Superresolution Diffuse Optical Imaging by Localization of Fluorescence

The multiple scattering of light presents major challenges in realizing useful in vivo imaging at tissue depths of more than about one millimeter, where many answers to health questions lie. Visible through near-infrared photons can be readily and safely detected through centimeters of tissue; however, limited information is available for image formation. One strategy for obtaining images is to model the photon transport and a simple incoherent model is the diffusion equation approximation to the Boltzmann transport equation. Such an approach provides a prediction of the mean intensity of heavily scattered light and hence provides a forward model for optimization-based computational imaging. While diffuse optical imaging methods have received substantial attention, they remain restricted in terms of resolution because of the loss of high-spatial-frequency information that is associated with the multiple scattering of photons. Consequently, only relatively large inhomogeneities, such as tumors or organs in small animals, can be effectively resolved. We introduce a superresolution imaging approach based on point localization in a diffusion framework that enables over two orders of magnitude improvement in the spatial resolution of diffuse optical imaging. The method is demonstrated experimentally by localizing a fluorescent inhomogeneity in a highly scattering slab and characterizing the localization uncertainty. The approach allows imaging through centimeters of tissue with a resolution of tens of microns, thereby enabling cells or cell clusters to be resolved. More generally, this high-resolution imaging approach could be applied with any physical transport or wave model and hence to a broad class of physical problems. Finally, paired with a suitable optical contrast mechanism, as can be realized with targeted fluorescent molecules or genetically modified animals, superresolution diffuse imaging should open alternative dimensions for in vivo applications.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2020

Deep medical image analysis with representation learning and neuromorphic computing

Deep learning is increasingly used in medical imaging, improving many steps of the processing chain, from acquisition to segmentation and anomaly detection to outcome prediction. Yet significant challenges remain: (i) image-based diagnosis depends on the spatial relationships between local patterns, something convolution and pooling often do not capture adequately; (ii) data augmentation, the de facto method for learning three-dimensional pose invariance, requires exponentially many points to achieve robust improvement; (iii) labelled medical images are much less abundant than unlabelled ones, especially for heterogeneous pathological cases; and (iv) scanning technologies such as magnetic resonance imaging can be slow and costly, generally without online learning abilities to focus on regions of clinical interest. To address these challenges, novel algorithmic and hardware approaches are needed for deep learning to reach its full potential in medical imaging.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

Gain characterization of LGAD sensors with beta particles and 28-MeV protons

Low Gain Avalanche Diodes, also known as LGADs, are widely considered for fast-timing applications in high energy physics, nuclear physics, space science, medical imaging, and precision measurements of rare processes. Such devices are silicon-based and feature an intrinsic gain due to a p + -doped layer that allows the production of a controlled avalanche of carriers, with multiplication on the order of 10–100. This technology can provide time resolution on the order of 20–30 ps, and variants of this technology can provide precision tracking too. The characterization of LGAD performance has so far primarily been focused on the interaction of minimum ionizing particles for high energy and nuclear physics applications. This article expands the study of LGAD performance to highly-ionizing particles, such as 28-MeV protons, which are relevant for several future scientific applications, e.g. in biology and medical physics, among others. These studies were performed with a beam of 28-MeV protons from a tandem Van de Graaff accelerator at Brookhaven National Laboratory and beta particles from a ^90Sr source; these were used to characterize the response and the gain of an LGAD as a function of bias voltage and collected charge. Here, the experimental results are also compared to TCAD simulations.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

Medical Imaging Equipment Energy Efficiency

The energy use of medical imaging equipment (MIE) in healthcare facilities is estimated to be about 5% of total site energy use. MIE energy use is only expected to increase as technology advances and more patients seek services from MIE, enabling more facilities to install additional MIE. Healthcare organizations need reliable information regarding comprehensive energy use or lifetime energy costs with which to make energy-efficient MIE procurement decisions, and there are currently no MIE energy or efficiency standards in the United States (U.S.). This fact sheet provides background on MIE energy consumption, key terms, and guidance on why and how to meter MIE, helping stakeholders make informed decisions to enhance their facility's performance.

ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION↗
DOE OSTI2023

Structural Changes to the Gd‐DTPA Complex at Varying Ligand Protonation State

Abstract Diethylenetriaminepentaacetic acid (DTPA) is a chelating agent whose complex with the Gd 3+ ion is used in medical imaging. DTPA is also used in lanthanide‐actinide separation processes. As protonation of the DTPA ligand can facilitate dissociation of the Gd 3+ ion from the Gd‐DTPA complex, this work investigates the coordination structures of the aqueous Gd 3+ ion and its environment when chelated by DTPA in eight different DTPA protonation states. Both classical and ab initio molecular dynamics (MD) simulations are conducted to model the solvated complexes. Extended X‐ray absorption fine structure (EXAFS) measurements of the Gd 3+ aqua ion, and the Gd‐DTPA complex at pH 1 and 11, are compared to EXAFS spectra predicted from the MD simulations to verify the accuracy of the MD structures. The findings of this work provide atomic‐level details into the fluctuating Gd‐DTPA complex environment as the DTPA ligand gradually detaches from the Gd 3+ ion with increased protonation.

Chemistry↗
DOE OSTI2018

Dunham Pad A 4H Well

Medical CT scans, raw and processed core logging data (including XRF) and core images of the Dunham Pad A 4H well, obtained by Triana Energy, as supplemental information to the NETL Technical Report Series Crandall, D.; Brown, S.; Moore, J.; Martin, K.; Mackey, P.; Carr, T. CT Scanning and Geophysical Measurements of the Marcellus Formation from the Dunham Pad A 4H Well NETL-TRS-X-2018; NETL Technical Report Series; U.S. Department of Energy, National Energy Technology Laboratory: Morgantown, WV, 2018; p 48.

Core Logs↗
DOE OSTI2022

Analysis of the laser welding keyhole using inline coherent imaging

Laser beam welding is a widely used fusion welding process in many industrial applications such as automotive, aerospace, energy, defense, and medical products. Industry has a fundamental need to model the laser welding process to minimize experimental testing and improve confidence in production welds. However, computational models attempting to predict weld formation are limited by an incomplete understanding of the beam-material interactions. As a consequence, these models do not accurately predict the mechanisms associated with laser weld formation. To improve the current state of prediction capabilities, it is vital to better detect/measure the physical aspects of the weld pool during high energy density welding. In this work, a novel real-time laser weld monitoring device using inline coherent imaging (ICI) was used to provide a fundamental understanding of laser weld formation via vaporization. The objective of this work was to investigate and quantify the relationship relating laser weld parameters and the vapor capillary (keyhole) through a state-of-the-art measurement technique. Bead-on-plate laser beam welds were produced with partial penetration on 304L stainless steel, 2205 duplex stainless steel, and Ti-6Al-4V. Keyhole monitoring was performed using a commercially available ICI system to collect keyhole penetration data in real time. These measurements were reconstructed to generate the vapor capillary shape at different welding parameters. Process parameters significantly influenced the keyhole shape and the keyhole root position relative to the process beam. Finally, the keyhole geometry showed distinct differences between the stainless steel alloys and Ti-6Al-4V.

2205↗
DOE OSTI2023

Health Hazards of Exposures to Radioiodine

Iodine is a chemical element with atomic number 53. Iodine-127 is stable (non-radioactive) and commonly found in nature. Elemental iodine is a purple-colored solid at room temperature and pressure, but spontaneously sublimates (turns into vapor). Iodine is an essential element for life, and is required for proper functioning of the thyroid. Iodine is present in many foods, and is readily absorbed by the body and concentrated in the thyroid gland. A fraction of iodine ingested or inhaled is rapidly removed by the kidneys. The rest of the inhaled or ingested iodine is absorbed the by thyroid and retained for many months. Iodine has a biological half-life of approximately 120 days in health individuals. The biological half-life can be shorter in individuals with hyperthyroidism, and longer in individuals with hypothyroidism. Iodine has a number of radioactive isotopes, most of which have relatively short half-lives (days or weeks). Short half-life iodine isotopes are useful for a variety of medical applications, including imaging and cancer therapy. For example, Iodine-123 (half-life 13 hours) is commonly used for medical imaging of the thyroid, while iodine-131 (half-life 8 days) is used for suppressing thyroid function in individuals with hyperthyroidism or ablating (killing) thyroid cells to treat thyroid cancer. Iodine-125 (half-life 59 days) is produced in nuclear reactors, and has medical uses. Although iodine-125 can be used for thyroid imaging, Iodine-123 is more commonly used for that purpose because of its shorter half-life and higher-energy emissions. Iodine-125 is more commonly used for cancer treatment, and can be processed into small metal pellets (seeds) inserted directly into a tumor. Iodine-125 emits low-energy x-rays which can kill tumor cells and generally cannot escape the tumor, sparing other tissues. Medical iodine for imaging or treatment is typically administered orally in the form a pill or liquid solution. A typical adult thyroid scan using iodine-123 involves having the patient swallow between one and four 0.1 millicuries pills, with the exact dose dependent on the patient’s weight. This results in a whole-body committed effective dose of 80 – 320 mrem, and a thyroid equivalent dose of 1443 – 5772 mrem. Note that the whole-body effective dose relates to the overall cancer risk, while the larger equivalent dose to the thyroid only indicates that most of this risk is the result of exposure to the thyroid. These doses are considered safe, although the procedure is not recommended for pregnant or breastfeeding women. In contrast, the quantities of iodine-131 used for treatment of hyperthyroidism and thyroid cancers are much higher. For treatment of hyperthyroidism, 4 – 10 millicuries are administered, while for thyroid cancer the administration can range from 50 – 150 millicuries of I-131. In addition to medical exposures, large populations were exposed to radioiodine as result of the atomic bombings of Hiroshima and Nagasaki in Japan, and the Chernobyl nuclear accident. These populations have been carefully followed for many years to assess the effect of their radiation exposures on cancer risk. As a result, a great deal is known about the cancer risks associated with radioiodine exposure. Because iodine is concentrated in the thyroid, the principal risk of exposure to radioiodine is thyroid cancer. Children have the highest risk of thyroid cancer after exposure to radioactive iodine. According to a large study of Japanese atomic bomb survivors, an effective dose of one Sievert (100,000 mrem) has been observed to increase the risk of thyroid cancer by a factor of 9.5 in children aged zero to nine years old, by a factor of 3 in children aged 10 to 19 years old, and by barely detectable amounts in adults. Another way of quantifying the risk from radioiodine exposure is from risk coefficients, which provide the risk per unit intake of radionuclides in terms of both morbidity (any cancer) and mortality (death). Both morbidity (risk of cancer) and mortality (death) risks are shown in the table below. Note that because thyroid cancer is almost never fatal, the morbidity coefficients are much larger than the mortality coefficients.

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2023

Reinforcement learning for real-time process control in high-temperature superconductor manufacturing

With high efficiency and low energy loss, high-temperature superconductors (HTS) have demonstrated their profound applications in various fields, such as medical imaging, transportation, accelerators, microwave devices, and power systems. The high-field applications of HTS tapes have raised the demand for producing cost-effective tapes with long lengths in superconductor manufacturing. However, achieving the uniform and enhanced performance of a long HTS tape is challenging due to the unstable growth conditions in the manufacturing process. Although it is confirmed that the process parameters during the advanced metal organic chemical vapor deposition (A-MOCVD) process influence the uniformity of the produced HTS tapes, the high-dimensional process parameter signals and their complicated interactions make it difficult to develop an effective control policy. In this paper, we propose a local measure for the uniformity of HTS tapes to provide instant feedback for our control policy. Then, we model the manufacturing of HTS tapes as a Markov decision process (MDP) with continuous state and action spaces to assess the instant reward in real time in our feedback control model. As our MDP involves continuous and high-dimensional state and action spaces, a neural fitted Q-iteration (NFQ) algorithm is adopted to solve the MDP with artificial neural network (ANN) function approximation. The collinearity of process parameters can restrict our capability of adjusting the process parameters, which is addressed by the principal component analysis (PCA) in our method. The control policy adjusts the PCA of process parameters using the NFQ algorithm. In conclusion, based on our case studies on real A-MOCVD dataset, the obtained control policy increases the average uniformity of tapes by 5.6% and performs especially well on sample HTS tapes with a low uniformity.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2019

Inverse Regression for Extraction of Tumor Site from Cancer Pathology Reports

Pathology reports are the primary source of information for cancer diagnosis of millions of the cancer patients across the United States. Cancer registries label these reports every year. The coded labels incorporate pertinent information such as cancer location, behavior, and histology. This information when combined with clinical information, medical imaging and even genomic information have a great potential to fuel discoveries in cancer research. The coding process is manual and requires many human experts to label the large volume of pathology reports in a timely manner. In this study, we have developed a supervised inverse regression based auto-labeler to automate the task. The experiments were conducted on a set of 942 pathology reports with human expert labels as the ground truth. We observed that the inverse regression based auto-labeler consistently performed better than or comparable to the best performing state-of-the-art method. These results demonstrate the potential of inverse regression for reliable information extraction from the pathology reports.

Dubey, Abhishek↗
DOE OSTI2025

Relativistic Laser Plasma Interactions At The Highest Intensities

High energy density science (HEDS) explores the nature of matter under extreme conditions of temperature and pressure. It is of fundamental importance and has many applications such as facilitating imaging with ions, neutrons, x-rays, and gamma rays with new applications being developed, including materials processing and medical therapies. In this project, we used high power, ultrashort pulse lasers to reach HEDS conditions. We have shown that low-cost, liquid crystal film based, double plasma mirror systems can be used to greatly improve laser pulse contrast while still maintaining high power and excellent spatial mode.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗