Search NASA⌕ Search

SEARCH · Search NASA

Results for “MRI”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
NASA NTRS2016

Initial Incidence of White Matter Hyperintensities on MRI in Astronauts

Introduction: Previous literature has described the increase in white matter hyperintensity (WMH) burden associated with hypobaric exposure in the U-2 and altitude chamber operating personnel. Although astronauts have similar hypobaric exposure pressures to the U2 pilot population, astronauts have far fewer exposures and each exposure would be associated with a much lower level of decompression stress due to rigorous countermeasures to prevent decompression sickness. Therefore, we postulated that the WMH burden in the astronaut population would be less than in U2 pilots. Methods: Twenty-one post-flight de-identified astronaut MRIs (5 mm slice thickness FLAIR sequences) were evaluated for WMH count and volume. The only additional data provided was an age range of the astronauts (43-57) and if they had ever performed an EVA (13 yes, 8 no). Results: WMH count in these 21 astronaut MRI was 21.0 +/- 24.8 (mean+/- SD) and volume was 0.382 +/- 0.602 ml, which was significantly higher than previously published results for the U2 pilots. No significant differences between EVA and no EVA groups existed. Age range of astronaut population is not directly comparable to the U2 population. Discussion: With significantly less frequent (sometimes none) and less stressful hypobaric exposures, yet a much higher incidence of increased WMH, this indicates the possibility of additional mechanisms beyond hypobaric exposure. This increase unlikely to be attributable just to the differences in age between astronauts and U2 pilots. Forward work includes continuing review of post-flight MRI and evaluation of pre to post flight MRI changes if available. Data mining for potential WMH risk factors includes collection of age, sex, spaceflight experience, EVA hours, other hypobaric exposures, hyperoxic exposures, radiation, high performance aircraft experience and past medical history. Finally, neurocognitive and vision/eye results will be evaluated for any evidence of impairment linked to increased WMH.

Norcross, Jason↗
DOE OSTI2024

MRI Load Profile Characterization Across Scanner Manufacturers and Field Strengths - Opportunities for Energy Efficiency Improvements

Radiology plays a vital role in patient care, but a reliance on energy-intensive imaging equipment leads to significant greenhouse gas emissions. MRI scanners can consume up to 80% of their total energy during times when there is not active patient imaging. Characterizing MRI average power and time to enter low-power mode across scanning protocols can identify energy efficiency opportunities.

emissions↗
DOE OSTI2026

Arm and shoulder muscle segmentation in axial MRI with UNet deep learning model

Quantifying individual upper-limb muscle volumes from MRI provides key insight into muscle-specific strength, deficits, and adaptations. Manual delineation is the gold standard but time‑intensive, and the performance of current deep learning approaches, particularly for small or anatomically complex muscles, remains incompletely characterized. We evaluated a state‑of‑the‑art deep learning framework across the entire upper limb and analyzed factors governing segmentation performance, with attention to the forearm. Three previously published MRI datasets (1.5 T, 3D GRE T1‑weighted; total n = 39) spanning young, middle‑aged, and older adults were curated and quality‑checked, including expert manual segmentations for 31 muscles. Following multiclass mask reconstruction, we trained three 3D nnU‑Net multiclass models matched to the muscle subsets present across datasets, using five‑fold cross‑validation and a composite Dice Similarity Coefficient (DSC) + cross entropy loss. Segmentation accuracy was assessed with DSC. Performance varied across muscles (mean DSC = 0.806 ± 0.098), ranging from 0.920 (Deltoid) to 0.461 (Extensor pollicis brevis). In uncertainty‑weighted regressions, muscle volume was positively associated with DSC (R2 = 0.36, p < 0.001), whereas training segmentation count and muscle orientation showed negligible associations (R2 ≤ 0.06). A weighted mixed‑effects model identified volume as the strongest evaluated predictor, explaining 23.9% of variance in DSC; orientation and training count each contributed <1%, leaving 61.5% unexplained. These results indicate that deep learning–based segmentation can accurately quantify muscle volume for many upper‑limb muscles but remains constrained for small, low‑contrast forearm muscles.

Gillespie, Samuel↗
DOE OSTI2025

Optimizing transmit field inhomogeneity of parallel RF transmit design in 7T MRI using deep learning

Ultrahigh field (UHF) Magnetic Resonance Imaging (MRI) provides a higher signal-to-noise ratio and, thereby, higher spatial resolution. However, UHF MRI introduces challenges such as transmit radiofrequency (RF) field (B+1) inhomogeneities, leading to uneven flip angles and image intensity anomalies. These issues can significantly degrade imaging quality and its medical applications. This study addresses B+1 field homogeneity through a novel deep learning-based strategy. Traditional methods like Magnitude Least Squares (MLS) optimization have been effective but are time-consuming and dependent on the patient’s presence. Recent machine learning approaches, such as RF Shim Prediction by Iteratively Projected Ridge Regression and deep learning frameworks, have shown promise but face limitations like extensive training times and oversimplified architectures. We propose a two-step deep learning strategy. First, we obtain the desired reference RF shimming weights from multi-channel B+1 fields using random-initialized Adaptive Moment Estimation. Then, we employ Residual Networks (ResNets) to train a model that maps B+1 fields to target RF shimming outputs. Our approach does not rely on pre-calculated reference optimizations for the testing process and efficiently learns residual functions. Comparative studies with traditional MLS optimization demonstrate our method’s advantages in terms of speed and accuracy. The proposed strategy achieves a faster and more efficient RF shimming design, significantly improving imaging quality at UHF. This advancement holds potential for broader applications in medical imaging and diagnostics.

Lu, Zhengyi [Vanderbilt University]↗
NASA NTRS1994

Simulation of arthroscopic surgery using MRI data

With the availability of Magnetic Resonance Imaging (MRI) technology in the medical field and the development of powerful graphics engines in the computer world the possibility now exists for the simulation of surgery using data obtained from an actual patient. This paper describes a surgical simulation system which will allow a physician or a medical student to practice surgery on a patient without ever entering an operating room. This could substantially lower the cost of medial training by providing an alternative to the use of cadavers. This project involves the use of volume data acquired by MRI which are converted to polygonal form using a corrected marching cubes algorithm. The data are then colored and a simulation of surface response based on springy structures is performed in real time. Control for the system is obtained through the use of an attached analog-to-digital unit. A remote electronic device is described which simulates an imaginary tool having features in common with both arthroscope and laparoscope.

Heller, Geoffrey↗
NASA NTRS1997

Predicting Bone Mechanical Properties of Cancellous Bone from DXA, MRI, and Fractal Dimensional Measurements

This project was aimed at making predictions of bone mechanical properties from non-invasive DXA and MRI measurements. Given the bone mechanical properties, stress calculations can be made to compare normal bone stresses to the stresses developed in exercise countermeasures against bone loss during space flight. These calculations in turn will be used to assess whether mechanical factors can explain bone loss in space. In this study we assessed the use of T2(sup *) MRI imaging, DXA, and fractal dimensional analysis to predict strength and stiffness in cancellous bone.

Harrigan, Timothy P.↗
NASA NTRS1998

MRI Evaluation of Spinal Length and Vertebral Body Angle During Loading with a Spinal Compression Harness

Weight bearing by the spinal column during upright posture often plays a role in the common problem of low back pain. Therefore, we developed a non-ferromagnetic spinal compression harness to enable MRI investigations of the spinal column during axial loading. Human subjects were fitted with a Nest and a footplate which were connected by adjustable straps to an analog load cell. MRI scans of human subjects (5 males and 1 female with age range of 27-53 yrs) during loaded and unloaded conditions were accomplished with a 1.5 Tesla GE Signa scanner. Studies of two subjects undergoing sequentially increasing spinal loads revealed significant decreases (r(sup 2) = 0.852) in spinal length between T4 and L5 culminating in a 1.5 to 2% length decrease during loading with 75% body weight. Sagittal vertebral body angles of four subjects placed under a constant 50% body weight load for one hour demonstrated increased lordotic and kyphotic curvatures. In the lumbar spine, the L2 vertebral body experienced the greatest angular change (-3 deg. to -5 deg.) in most subjects while in the thoracic spine, T4 angles increased from the unloaded state by +2 deg. to +9 deg. Overall, our studies demonstrate: 1) a progressive, although surprisingly small, decrease in spinal length with increasing load and 2) relatively large changes in spinal column angulation with 50% body weight.

Campbell, James A.↗
NASA NTRS1998

Low-field MRI of laser polarized noble gas

NMR images of laser polarized 3He gas were obtained at 21 G using a simple, homebuilt instrument. At such low fields magnetic resonance imaging (MRI) of thermally polarized samples (e.g., water) is not practical. Low-field noble gas MRI has novel scientific, engineering, and medical applications. Examples include portable systems for diagnosis of lung disease, as well as imaging of voids in porous media and within metallic systems.

Non-NASA Center↗
NASA NTRS1987

See-saw nystagmus and brainstem infarction: MRI findings

A patient with see-saw nystagmus had a lesion localized by Magnetic Resonance Imaging (MRI) to the paramedian ventral midbrain with involvement of the right interstitial nucleus of Cajal. This the first MRI study of see-saw nystagmus associated with a presumed brainstem vascular event. Our findings support animal and human studies suggesting that dysfunction of the interstitial nucleus of Cajal or its connections is central in this disorder.

Non-NASA Center↗
DOE OSTI2026

A Centralized AI Lakehouse Framework for Brain Tumor MRI Classification and Segmentation, University KPI Forecasting, and Water Potability Prediction

In many university and healthcare projects, models are built for very different data types such as tables, institutional time series, and medical images, but they are deployed as separate applications. In this work, that separation made testing and maintenance difficult because each module had its own pipeline and runtime requirements. This paper presents an integrated AI lakehouse-style implementation that runs three model pipelines inside one containerized backend. For medical imaging, we used MRI datasets from IEEE DataPort: a four-class classification set with 7012 images (5708 train/1304 test) and a segmentation set with 3063 image–mask pairs. The classification model (ResNet50 transfer learning) is evaluated using a proper train–validation–test protocol across multiple splits (80/10/10, 70/10/20, 60/10/30, and 10/30/60), achieving a test accuracy of 99.00% under the standard 80/10/10 split. Additionally, a patient-level evaluation is conducted using an external glioma dataset to provide a more realistic assessment without data leakage. The segmentation model (DeepLabV3-ResNet50) achieved 83.09% validation mIoU and 88.79% Dice score. For university KPI forecasting, we used annual IPEDS and NSF HERD data from 2010 to 2023 for three universities (BSU, EOU, and UAB). To examine the effect of preprocessing on forecasting performance, two case studies are conducted. In the first case, linear interpolation is applied to generate semester-level data. In the second case, the original annual data is used directly without interpolation. Random Forest regression and ARIMA models are evaluated using MAE, RMSE, MAPE, and R 2 . The results showed that interpolation improved apparent forecasting performance due to smoothing, while evaluation on the original annual data provided a more realistic assessment of model behavior. To further validate the framework on a larger dataset, an additional case study is conducted using a student dropout dataset. For water potability, we trained and compared multiple tabular classifiers on a large dataset (1,048,575 samples). A Random Forest model (100 trees, max depth 10) achieved 85.86% test accuracy and high recall for unsafe samples (0.8447). All modules are served via FastAPI and deployed together using Docker, with workflow automation routing requests to the correct endpoint. System-level benchmarking indicates that the backend maintains stable throughput and latency under concurrent requests.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Fast-RF-Shimming: Accelerate RF shimming in 7T MRI using deep learning

Ultrahigh field (UHF) Magnetic Resonance Imaging (MRI) offers an elevated signal-to-noise ratio (SNR), enabling exceptionally high spatial resolution that benefits both clinical diagnostics and advanced research. However, the jump to higher fields introduces complications, particularly transmit radiofrequency (RF) field ($B^{+}_{1}$) inhomogeneities, manifesting as uneven flip angles and image intensity irregularities. These artifacts can degrade image quality and impede broader clinical adoption. Traditional RF shimming methods, such as Magnitude Least Squares (MLS) optimization, effectively mitigate $B^{+}_{1}$ inhomogeneity, but remain time-consuming. Recent machine learning approaches, including RF Shim Prediction by Iteratively Projected Ridge Regression and other deep learning architectures, suggest alternative pathways. Although these approaches show promise, challenges such as extensive training periods, limited network complexity, and practical data requirements persist. In this paper, we introduce a holistic learning-based framework called Fast-RF-Shimming, which achieves a 5000 ​× ​speed-up compared to the traditional MLS method. In the initial phase, we employ random-initialized Adaptive Moment Estimation (Adam) to derive the desired reference shimming weights from multi-channel $B^{+}_{1}$ fields. Next, we train a Residual Network (ResNet) to map $B^{+}_{1}$ fields directly to the ultimate RF shimming outputs, incorporating the confidence parameter into its loss function. Finally, we design Non-uniformity Field Detector (NFD), an optional post-processing step, to ensure the extreme non-uniform outcomes are identified. Comparative evaluations with standard MLS optimization underscore notable gains in both processing speed and predictive accuracy, which indicates that our technique shows a promising solution for addressing persistent inhomogeneity challenges.

Deep learning↗
DOE OSTI2025

PHASE: Personalized Head-based Automatic Simulation for Electromagnetic properties in 7T MRI

Accurate and individualized human head models are becoming increasingly important for electromagnetic (EM) simulations. These simulations depend on precise anatomical representations to realistically model electric and magnetic field distributions, particularly when evaluating Specific Absorption Rate (SAR) within safety guidelines. State of the art simulations use the Virtual Population due to limited public resources and the impracticality of manually annotating patient data at scale. Here, this paper introduces Personalized Head-based Automatic Simulation for EM properties (PHASE), an automated open-source toolbox that generates high-resolution, patient-specific head models for EM simulations using paired T1-weighted (T1w) magnetic resonance imaging (MRI) and computed tomography (CT) scans with 14 tissue labels. To evaluate the performance of PHASE models, we conduct semi-automated segmentation and EM simulations on 15 real human patients, serving as the gold standard reference. The PHASE model achieved comparable global SAR and localized SAR averaged over 10 grams of tissue (SAR-10g), demonstrating its potential as a promising tool for generating large-scale human model datasets in the future. The code and models of PHASE toolbox have been made publicly available: https://github.com/hrlblab/PHASE.

Deep learning↗
DOE OSTI2025

Using 3D Measurements from MRI to Improve VVUQ

Magnetic resonance velocimetry (MRV) is a diagnostic that can measure 3D, three-component turbulent velocity fields in arbitrarily complex flow configurations. The approach uses magnetic resonance imaging (MRI), which is a system commonly available in radiology departments or medical research centers. MRV acquires signal from hydrogen protons in water channel flows. Despite the obvious utility of the measurements for investigating a variety of flows, comparisons with computational fluid dynamic (CFD) simulations largely retain traditional metrics, comparing velocity profiles and planes at discrete regions within the flow rather than using the 3D nature of the measured field. In this effort, MRV was conducted in a simple water channel with six spanwise-centered periodic obstacles roughly shaped as cubes. At a fully turbulent Reynolds number of 15,000, the channel includes two flows: a main streamwise flow and, in the wake of the second obstacle, an injected flow oriented perpendicular to the streamwise flow. The flow geometry includes partial obstacles on the side walls and is inherently 3D because of the interaction of the two streams and the wake features, making the flow challenging to completely measure with traditional optical techniques. Aside from line profiles and planar comparisons, two 3D metrics are used to compare the data with a steady Reynolds-averaged Navier–Stokes (RANS) simulation result. A brief discussion about the comparison is provided, including comments about uncertainty.

Benson, Mike [ORNL] (ORCID:000000023210116X)↗
NASA NTRS1992

Magnetic Resonance Imaging (MRI) of skeletal muscles in astronauts after 9 days of space flight

Skylab data indicated that prolonged exposure of human subjects to microgravity environment causes significant muscle atrophy accompanied by reduced muscle strength and fatigue resistance. The objective of this study was to determine decrements in muscle size, if any, in the soleus and gastrocnemius muscles of male and female astronauts after 9 days of space flight. Methods: Eight astronauts, one female and seven male, between the ages of 31 and 59 years 59-84 kg in body weight were examined by MRI 2-3 times preflight within 16 days before launch, and 2 days, (n=6) and seven days (n=3) after landing. The right leg muscles (gastroc-soleus) were imaged with a lower extremity coil in magnets operating at 1.0 or 1.5 Tsela. The imaging protocol consisted of spin echo with a Tr of 0.70 - 1.5 sec. Thirty to forty 3-5 mm thick slices were acquired in 256 x 128 or 256 x 256 matrices. Acquisition time lasted 20-40 minutes. Multiple slices were measured by computerized planimetry. Results: Compared to the preflight, the cross-sectoral areas (CSA) of the soleus, gastrocnemius, and the leg, at 2 days after landing were reduced (at least p less than 0.05) 8.9 percent, 13.2 percent, and 9.5 percent respectively. The soleus and the leg of three astronauts evaluated at 7 days postflight did not show full recovery compared to the preflight values. Conclusions: It is concluded that l9-days of space flight may cause significant decreases in CSA of the leg muscles. The factors responsible for this loss need further determination.

Jaweed, M.↗
NASA NTRS1997

Development of Human Muscle Protein Measurement with MRI

It is known that micro-gravity has a strong influence on the human musculoskeletal system. A number of studies have shown that significant changes in skeletal muscles occur in both space flight and bedrest simulation. In our 5 week bedrest study, the cross-sectional area of soleus-gastrocnemius decreased about 12% while the cross-sectional area of anterior calf muscles decreased about 4%. Using volume measurements, these losses increased after 17 weeks to approximately 30% and 21% respectively. Significant muscle atrophy was also found on the SL-J crew members after only 8 days in space. It is important that these effects are fully understood so that countermeasures can be developed. The same knowledge might also be useful in preventing muscle atrophy related to other medical problems. A major problem with anatomical measurements of muscle during bed rest and microgravity is the influence of fluid shifts and water balance on the measurement of muscle volume, especially when the exposure duration is short and the atrophy is relatively small. Fluid shifts were documented in Skylab by visual observations of blood vessel distention, rapid changes in limb volume, center of mass measurements and subjective descriptions such as puffy faces and head fullness. It has been reported that the muscle water content of biopsied soleus muscles decreased following 8 hours of head down tilt bed rest. Three aspects of fluid shifts that can affect volume measurements are: first, the shift of fluid that occurs whenever there is a change from upright to a recumbent position and vice versa; second, the potential for fluid accumulation in the lower limbs resulting from muscle damage caused by overextending atrophied muscle or swelling caused by deconditioned precapillary sphincter muscles during reambulation; third, the net change of hydration level during and after bed rest or spaceflight. Because of these transitory fluid shifts, muscle protein is expected to represent muscle capacity better than does muscle volume. The purpose of this study is to test the feasibility of using MRI to quantify of muscle protein and water content changes in muscle.

Lin, Chen↗
NASA NTRS1999

MRI Measurements and Granular Dynamics Simulation of Segregation of Granular Mixture

A counter intuitive axial segregation phenomenon in a rotating horizontal cylinder has recently captured attention of many researchers in different disciplines. There is a growing consensus that the interplay between the particle dynamics and the evolution of the internal structure during the segregation process must be carefully investigated. Magnetic resonance imaging (MRI) has been used to non-invasively obtain much needed dynamic/static information such as velocity and concentration profiles, and it has proven to be capable of depicting the evolution of segregation processes. Segregation in a rotating cylinder involves two processes: the first is to transport small particles in the radial direction to form a radial core, and the second is to transform the radial core into axially segregated bands. Percolation and/or "stopping" have been proposed as mechanisms for the radial segregation. As to mechanisms for axial band formation, much less is known. The difference in the dynamic angle of repose has been proposed to segregate different components in the axial direction. Recently, Hill and Kakalios have reported that particles mix or demix depending upon the competition between diffusion and preferential drift whose order can be determined by the dynamic angle of repose through the adjustment of the rotation rate. We claim that the dynamic angle of repose could be one of the causes, however, it fails to offer reasonable explanations for certain aspects of the axial migration. For example, we always observe that the radial segregation precedes the axial segregation and small particles migrate in the radial direction to form an axially extended radial core. It then transforms into axially segregated bands. By definition, the effects of the dynamic angle of repose are restricted near the free surface where the flowing layer is present. However, during the process of transforming from the radially segregated core to axially segregated bands, small particles located in the deep core region, which is untouched by the flowing layer, also completely disappear. Usually, the dynamics angle of repose are uniquely defined for individual species to characterize particle properties, and the dynamic angle of repose thus defined provides little information for the dynamic angle of repose of the mixture since the concentration ratio and the internal packing structure do not remain the same during the segregation processes. Under microgravity environment, the dynamics angle of repose argument does not hold since there is simply no flowing layer to influence/determine the preferred directions of segregation. We have thus designed an experiment so that the effects of the dynamic angle of repose can be minimized by filling the cylinder almost completely full. Small particles still formed a radial core and also migrated to form axial bands. As ground based experiments we have designed and conducted both 2D and 3D segregation experiments. The 2D experiments are performed using a thin cylinder (the gap between two end caps is about 5 mm) filled with different combinations of particles. The 3D experiments are conducted with a long cylinder of its length and diameter of 27cm and 7cm, respectively. Results of 2D experiments indicate that different mechanisms govern particle motion in regions near and far from the axis of rotation. Results of 3D experiments indicate that a series of collapses of microstructures of particle packing (micro-collapses) may be responsible for the creation of voids for small particles to migrate through in the axial direction. We have successfully eliminated the dynamic angle of repose as a cause for segregation, however, by almost completely filling the cylinder with the particles, we have lost an opportunity to investigate a possibility of particle "mobility" being a cause for segregation which requires a flowing surface but not the difference in the angle of repose. This is currently being investigated.

Nakagawa, M.↗
NASA NTRS2004

Estimation of Length-Scales in Soils by MRI

Soil can be best described as an unconsolidated granular media that forms porous structure. The present macroscopic theory of water transport in porous media rests upon the continuum hypothesis that the physical properties of porous media can be associated with continuous, twice-differentiable field variables whose spatial domain is a set of centroids of Representative Elementary Volume (REV) elements. MRI is an ideal technique to estimate various length-scales in porous media. A 0.267 T permanent magnet at NASA GRC was used for this study. A 2D or 3D spatially-resolved porosity distribution were obtained from the NMR signal strength from each voxel and the spin-lattice relaxation time. A classical spin-warp imaging with Multiple Spin Echos (MSE) was used to evaluate proton density in each voxel. Initial resolution of 256 x 256 was subsequently reduced by averaging neighboring voxels and the porosity convergence was observed. A number of engineered "space candidate" soils such as Isolite(trademark), Zeoponics(trademark), Turface(trademark), and Profile(trademark) were used. Glass beads in the size range between 50 microns to 2 mm were used as well. Initial results with saturated porous samples have shown a good estimate of the average porosity consistent with the gravimetric porosity measurement results. For Profile(trademark) samples with particle sizes ranging between 0.25 to 1 mm and characteristic interparticle pore size of 100 microns the characteristic Darcy scale was estimated to be about delta(sub REV) = 10 mm. Glass beads porosity show clear convergence toward a definite REV which stays constant throughout homogeneous sample. Additional information is included in the original extended abstract.

N E Daidzic↗
NASA NTRS2009

Gradiometer Using Middle Loops as Sensing Elements in a Low-Field SQUID MRI System

A new gradiometer scheme uses middle loops as sensing elements in lowfield superconducting quantum interference device (SQUID) magnetic resonance imaging (MRI). This design of a second order gradiometer increases its sensitivity and makes it more uniform, compared to the conventional side loop sensing scheme with a comparable matching SQUID. The space between the two middle loops becomes the imaging volume with the enclosing cryostat built accordingly.

Penanen, Konstantin↗