Search NASA⌕ Search

SEARCH · Search NASA

Results for “imaging methods”

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 1,207 records · Page 67
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

Utilizing broadband X-rays in a Bragg coherent X-ray diffraction imaging experiment

A method is presented to simplify Bragg coherent X-ray diffraction imaging studies of complex heterogeneous crystalline materials with a two-stage screening/imaging process that utilizes polychromatic and monochromatic coherent X-rays and is compatible with in situ sample environments. Coherent white-beam diffraction is used to identify an individual crystal particle or grain that displays desired properties within a larger population. A three-dimensional reciprocal-space map suitable for diffraction imaging is then measured for the Bragg peak of interest using a monochromatic beam energy scan that requires no sample motion, thus simplifying in situ chamber design. This approach was demonstrated with Au nanoparticles and will enable, for example, individual grains in a polycrystalline material of specific orientation to be selected, then imaged in three dimensions while under load.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Characterizing and Mitigating Intraday Variability: Reconstructing Source Structure in Accreting Black Holes with mm-VLBI

The extraordinary physical resolution afforded by the Event Horizon Telescope has opened a window onto the astrophysical phenomena unfolding on horizon scales in two known black holes, M87* and Sgr A*. However, with this leap in resolution has come a new set of practical complications. Sgr A* exhibits intraday variability that violates the assumptions underlying Earth aperture synthesis, limiting traditional image reconstruction methods to short timescales and data sets with very sparse (u, v) coverage. We present a new set of tools to detect and mitigate this variability. We develop a data-driven, model-agnostic procedure to detect and characterize the spatial structure of intraday variability. This method is calibrated against a large set of mock data sets, producing an empirical estimator of the spatial power spectrum of the brightness fluctuations. We present a novel Bayesian noise modeling algorithm that simultaneously reconstructs an average image and statistical measure of the fluctuations about it using a parameterized form for the excess variance in the complex visibilities not otherwise explained by the statistical errors. These methods are validated using a variety of simulated data, including general relativistic magnetohydrodynamic simulations appropriate for Sgr A* and M87*. We find that the reconstructed source structure and variability are robust to changes in the underlying image model. We apply these methods to the 2017 EHT observations of M87*, finding evidence for variability across the EHT observing campaign. The variability mitigation strategies presented are widely applicable to very long baseline interferometry observations of variable sources generally, for which they provide a data-informed averaging procedure and natural characterization of inter-epoch image consistency.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

A new method of relative radiometric calibration for hyperspectral imaging based on skylight monitor

A new method of relative radiometric calibration based on the skylight monitor is presented for airborne hyperspectral remote sensing to eliminate the negative effect on data resulting from the time-varying skylight during a long flight. In terms of large area operation of mapping or spectral imaging, a static radiation model hardly reveals actual cases as external environment keeps changing. Compared with common considerable factors, fluctuation of skylight is the primary one that may observably affect spectral imaging continuously leading to radiation changes upon flight lines. To get rid of the influence of skylight on spectral imagery and monitoring the variation during the flight, we assemble a set of skylight monitor fixed to the aircraft cabin for the sake of recording the semi-spherical skylight keeping synchronization with hyperspectral imaging system and the position and orientation system. The skylight monitor consists of a cosine corrector on the top receiving the skylight, a stable optical fiber and a visible and near-infrared linear array modular spectrometer. The corresponding matched data of skylight with each frame of images is used to correct related digital number after stripe non-uniformity correction, achieving mean relative standard deviation (coefficient of variation or CV) of under 6% for all 256 bands. The final result shows a satisfactory calibrated image of an appropriate evaluation criteria with contrast of the original image.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2017

Simulation of Laser Ultrasonics for Detection of Surface-Connected Rail Defects

Laser ultrasonic produces frequencies in the MHz range, enabling high accuracy and a strong ability to detect rail surface defects. This paper mainly studied on the simulation of detecting surface-connected rail defects on 60 kg rails with laser ultrasonic, established the finite element model of laser-excited ultrasonic Rayleigh wave, carried out the simulation, and verified the effectiveness of the technology through experiments. To solve the problem that laser ultrasonic is insensitive to the width of defects in actual detection, and unable to make quantitative detection of defects, this paper established a new model on the basis of improving the original model that has been verified, exciting ultrasonic at the two sides at the same time of a rail with two staggered beams of laser separately to detect irregular scratch defects on rail surface, and two groups of signal data were received through two probes. Each group of data can present the half-profile information of defects, and further form two detection images of the defect. At last, the two detection images were combined into a complete image through image processing. The results of the experiment indicate that the technology studied offers a new method for the effective quantitative detection of surface-connected defects on rail.

42 ENGINEERING↗
DOE OSTI2015

WE-G-BRD-07: Automated MR Image Standardization and Auto-Contouring Strategy for MRI-Based Adaptive Brachytherapy for Cervix Cancer

Purpose: MRI-based adaptive brachytherapy (ABT) is an emerging treatment modality for patients with gynecological tumors. However, MR image intensity non-uniformities (IINU) can vary from fraction to fraction, complicating image interpretation and auto-contouring accuracy. We demonstrate here an automated MR image standardization and auto-contouring strategy for MRI-based ABT of cervix cancer. Methods: MR image standardization consisted of: 1) IINU correction using the MNI N3 algorithm, 2) noise filtering using anisotropic diffusion, and 3) signal intensity normalization using the volumetric median. This post-processing chain was implemented as a series of custom Matlab and Java extensions in MIM (v6.4.5, MIM Software) and was applied to 3D T2 SPACE images of six patients undergoing MRI-based ABT at 3T. Coefficients of variation (CV=σ/µ) were calculated for both original and standardized images and compared using Mann-Whitney tests. Patient-specific cumulative MR atlases of bladder, rectum, and sigmoid contours were constructed throughout ABT, using original and standardized MR images from all previous ABT fractions. Auto-contouring was performed in MIM two ways: 1) best-match of one atlas image to the daily MR image, 2) multi-match of all previous fraction atlas images to the daily MR image. Dice’s Similarity Coefficients (DSCs) were calculated for auto-generated contours relative to reference contours for both original and standardized MR images and compared using Mann-Whitney tests. Results: Significant improvements in CV were detected following MR image standardization (p=0.0043), demonstrating an improvement in MR image uniformity. DSCs consistently increased for auto-contoured bladder, rectum, and sigmoid following MR image standardization, with the highest DSCs detected when the combination of MR image standardization and multi-match cumulative atlas-based auto-contouring was utilized. Conclusion: MR image standardization significantly improves MR image uniformity. The combination of MR image standardization and multi-match cumulative atlas-based auto-contouring produced the highest DSCs and is a promising strategy for MRI-based ABT for cervix cancer.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

SU-F-J-20: Commissioning and Acceptance Testing of the C-Rad CatalystHD Surface Imaging System

Introduction: With the increasing use of surface-based, nonionizing image-guided radiotherapy (IGRT) systems, a comprehensive set of clinical acceptance and commissioning procedures are needed to ensure correct functionality and proper clinical integration. Although TG-147 provides a specific set of parameters, measurement methodologies have yet to be described. The aim of this study was to provide a comprehensive overview of the commissioning and acceptance analysis performed for the C-Rad CatalystHD imaging system. Methods and Materials: Methodology for the commissioning and acceptance of the C-Rad CatalystHD imaging system was developed using commercially available clinical equipment. Following TG-147 guidelines, the following tests were performed: integration of peripheral equipment, system drift, static spatial reproducibility and localization accuracy, static end-to-end analysis, static rotational accuracy, dynamic spatial accuracy, dynamic temporal accuracy, dynamic radiation delivery and a comprehensive end-to-end analysis. Results: The field of view (FOV) of the CatalystHD was 105×109×83 cm3 in the lateral, longitudinal and vertical directions. For thermal equilibrium and system drift, a thermal drift of 1.0mm was noted. A 45 min warmup time is recommended if the system has been shut off an extended period of time (>24 hours) before the QA procedure to eliminate any thermal drift. Spatial reproducibility was found to be 0.05±0.03 mm using a rigid phantom. For the static localization accuracy, system agreement with couch shifts was within 0.1±0.1 mm and positioning agreement with kV-CBCT was 0.16±0.10 mm. For static rotational accuracy, system agreement with a high precision rotational stage (0.01 deg precision) was within 0.10±0.07 deg. Dynamic spatial and temporal localization accuracy was found to be within 0.2±0.1 mm. Conclusion: A comprehensive commissioning and acceptance study was performed using commercially available phantoms and in-house methodologies to provide a performance evaluation of the CatalystHD imaging system.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

Single-view phase retrieval of an extended sample by exploiting edge detection and sparsity

We propose a new approach to robustly retrieve the exit wave of an extended sample from its coherent diffraction pattern by exploiting sparsity of the sample's edges. This approach enables imaging of an extended sample with a single view, without ptychography. We introduce nonlinear optimization methods that promote sparsity, and we derive update rules to robustly recover the sample's exit wave. We test these methods on simulated samples by varying the sparsity of the edge-detected representation of the exit wave. Finally, our tests illustrate the strengths and limitations of the proposed method in imaging extended samples.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2013

High Fidelity Computational Analysis of CO2 Trapping at Pore Scales (Final Technical Report)

With an alarming rise in carbon dioxide (CO2) emission from anthropogenic sources, CO2 sequestration has become an attractive choice to mitigate the emission. Some popular storage media for CO 2 are oil reservoirs, deep coal-bed, and deep oceanic-beds. These have been used for the long term CO 2 storage. Due to special lowering viscosity and surface tension property of CO 2 , it has been widely used for enhanced oil recovery. The sites for CO 2 sequestration or enhanced oil recovery mostly consist of porous rocks. Lack of knowledge of molecular mobility under confinement and molecule-surface interactions between CO2 and natural porous media results in generally governed by unpredictable absorption kinetics and total absorption capacity for injected fluids, and therefore, constitutes barriers to the deployment of this technology. Therefore, it is important to understand the flow dynamics of CO 2 through the porous microstructures at the finest scale (pore-scale) to accurately predict the storage potential and long-term dynamics of the sequestered CO 2 . This report discusses about pore-network flow modeling approach using variational method and analyzes simulated results this method simulations at pore-scales for idealized network and using Berea Sandstone CT scanned images. Variational method provides a promising way to study the kinetic behavior and storage potential at the pore scale in the presence of other phases. The current study validates variational solutions for single and two-phase Newtonian and single phase non-Newtonian flow through angular pores for special geometries whose analytical and/or empirical solutions are known. The hydraulic conductance for single phase flow through a triangular duct was also validated against empirical results derived from lubricant theory.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2014

Characterization of a commercial hybrid iterative and model-based reconstruction algorithm in radiation oncology

Purpose: Iterative reconstruction (IR) reduces noise, thereby allowing dose reduction in computed tomography (CT) while maintaining comparable image quality to filtered back-projection (FBP). This study sought to characterize image quality metrics, delineation, dosimetric assessment, and other aspects necessary to integrate IR into treatment planning. Methods: CT images (Brilliance Big Bore v3.6, Philips Healthcare) were acquired of several phantoms using 120 kVp and 25–800 mAs. IR was applied at levels corresponding to noise reduction of 0.89–0.55 with respect to FBP. Noise power spectrum (NPS) analysis was used to characterize noise magnitude and texture. CT to electron density (CT-ED) curves were generated over all IR levels. Uniformity as well as spatial and low contrast resolution were quantified using a CATPHAN phantom. Task specific modulation transfer functions (MTF{sub task}) were developed to characterize spatial frequency across objects of varied contrast. A prospective dose reduction study was conducted for 14 patients undergoing interfraction CT scans for high-dose rate brachytherapy. Three physicians performed image quality assessment using a six-point grading scale between the normal-dose FBP (reference), low-dose FBP, and low-dose IR scans for the following metrics: image noise, detectability of the vaginal cuff/bladder interface, spatial resolution, texture, segmentation confidence, and overall image quality. Contouring differences between FBP and IR were quantified for the bladder and rectum via overlap indices (OI) and Dice similarity coefficients (DSC). Line profile and region of interest analyses quantified noise and boundary changes. For two subjects, the impact of IR on external beam dose calculation was assessed via gamma analysis and changes in digitally reconstructed radiographs (DRRs) were quantified. Results: NPS showed large reduction in noise magnitude (50%), and a slight spatial frequency shift (∼0.1 mm{sup −1}) with application of IR at L6. No appreciable changes were observed for CT-ED curves between FBP and IR levels [maximum difference ∼13 HU for bone (∼1% difference)]. For uniformity, differences were ∼1 HU between FBP and IR. Spatial resolution was well conserved; the largest MTF{sub task} decrease between FBP and IR levels was 0.08 A.U. No notable changes in low-contrast detectability were observed and CNR increased substantially with IR. For the patient study, qualitative image grading showed low-dose IR was equivalent to or slightly worse than normal dose FBP, and is superior to low-dose FBP (p < 0.001 for noise), although these did not translate to differences in CT number, contouring ability, or dose calculation. The largest CT number discrepancy from FBP occurred at a bone/tissue interface using the most aggressive IR level [−1.2 ± 4.9 HU (range: −17.6–12.5 HU)]. No clinically significant contour differences were found between IR and FBP, with OIs and DSCs ranging from 0.85 to 0.95. Negligible changes in dose calculation were observed. DRRs preserved anatomical detail with <2% difference in intensity from FBP combined with aggressive IRL6. Conclusions: These results support integrating IR into treatment planning. While slight degradation in edges and shift in texture were observed in phantom, patient results show qualitative image grading, contouring ability, and dosimetric parameters were not adversely affected.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2017

Phantom Study Investigating the Accuracy of Manual and Automatic Image Fusion with the GE Logiq E9: Implications for use in Percutaneous Liver Interventions

PurposeTo determine the accuracy of automatic and manual co-registration methods for image fusion of three-dimensional computed tomography (CT) with real-time ultrasonography (US) for image-guided liver interventions.Materials and MethodsCT images of a skills phantom with liver lesions were acquired and co-registered to US using GE Logiq E9 navigation software. Manual co-registration was compared to automatic and semiautomatic co-registration using an active tracker. Also, manual point registration was compared to plane registration with and without an additional translation point. Finally, comparison was made between manual and automatic selection of reference points. In each experiment, accuracy of the co-registration method was determined by measurement of the residual displacement in phantom lesions by two independent observers.ResultsMean displacements for a superficial and deep liver lesion were comparable after manual and semiautomatic co-registration: 2.4 and 2.0 mm versus 2.0 and 2.5 mm, respectively. Both methods were significantly better than automatic co-registration: 5.9 and 5.2 mm residual displacement (p < 0.001; p < 0.01). The accuracy of manual point registration was higher than that of plane registration, the latter being heavily dependent on accurate matching of axial CT and US images by the operator. Automatic reference point selection resulted in significantly lower registration accuracy compared to manual point selection despite lower root-mean-square deviation (RMSD) values.ConclusionThe accuracy of manual and semiautomatic co-registration is better than that of automatic co-registration. For manual co-registration using a plane, choosing the correct plane orientation is an essential first step in the registration process. Automatic reference point selection based on RMSD values is error-prone.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2015

SU-E-T-453: Image-Guided Patient-Specific Quality Assurance for Multi-Target Single Isocenter Intracranial VMAT Radiosurgery

Purpose: To demonstrate a patient specific, image-guided quality assurance method that tests both dosimetric and geometric accuracy for single-isocenter multiple-target VMAT radiosurgery (SIMT-VMAT-SRS) Method: We used a new film type, EBT-XD (optimal range 0.4–40Gy), and an in-house PMMA phantom having a coronal plane for film and a 0.125 cm3 ionization chamber (IC). The phantom contained fiducial features for kV image guided setup and for accurate film marking. Five patient plans with multiple targets sizes ranging from 3 to 21mm in diameter and prescribed doses from 14 to 18 Gy were selected. Two verification plans were generated for each case with the film plane passing through the center of the largest and smallest targets. For the four largest targets we obtained an IC measurement. For each case, a calibration film was irradiated using a custom designed step pattern. The films were scanned using a flatbed color scanner and converted to dose using the calibration film and the three channel calibration method. Image registration was performed between film and treatment planning system calculations to evaluate the geometric accuracy. Results: The mean registration vector had an average magnitude of 0.47 mm (range from 0.13mm to 0.64 mm). For the four largest targets, the mean ratio of the IC and film measurement to expected dose was 0.990 (range 0.968 to 1.009) and 1.032 (1.021 to 1.046), respectively. The fraction of pixels having gamma index < 1 for criteria of 3%/3mm, 3%/2mm, 3%/1mm was 98.8%, 97.5% and 87.2% before geometric registration and 99.1%, 98.3% and 94.8% after registration. Conclusion: We have demonstrated an image-guided QA method can assess both geometric and dosimetric accuracy. The phantom was positioned with sub-millimeter accuracy. Absolute film dosimetry using EBT-XD film was sufficiently accurate for assessment of dose to multi-targets too small for IC measurement in SRS VMAT plans.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2018

Method for phase retrieval to reduce a sampling requirement when imaging a dynamic process

A method for retrieving phase information in a coherent diffraction imaging process includes acquiring a plurality of 3D data sets, each 3D data set corresponding to one of a plurality of time states, and reconstructing a 3D image of the object at a given time state using the 3D data set from all of the time states. Each 3D data set is acquired by: illuminating an object positioned in a first position with a coherent beam; measuring a first 2D diffraction pattern using an area detector; rotating the object around a tilt axis thereof to a second position that is different from the first position; re-illuminating the object positioned in the second position with the coherent beam; re-measuring a second 2D diffraction pattern using the area detector; and repeating the rotating, re-illuminating and re-measuring steps such that each 3D data set includes a predetermined number of diffraction patterns.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Predicting Nugget Size of Resistance Spot Welds Using Infrared Thermal Videos With Image Segmentation and Convolutional Neural Network

Resistance spot welding (RSW) is a widely adopted joining technique in automotive industry. Recent advancement in sensing technology makes it possible to collect thermal videos of the weld nugget during RSW using an infrared (IR) camera. The effective and timely analysis of such thermal videos has the potential of enabling in situ nondestructive evaluation (NDE) of the weld nugget by predicting nugget thickness and diameter. Deep learning (DL) has demonstrated to be effective in analyzing imaging data in many applications. However, the thermal videos in RSW present unique data-level challenges that compromise the effectiveness of most pre-trained DL models. We propose a novel image segmentation method for handling the RSW thermal videos to improve the prediction performance of DL models in RSW. The proposed method transforms raw thermal videos into spatial-temporal instances in four steps: video-wise normalization, removal of uninformative images, watershed segmentation, and spatial-temporal instance construction. The extracted spatial-temporal instances serve as the input data for training a DL-based NDE model. The proposed method is able to extract high-quality data with spatial-temporal correlations in the thermal videos, while being robust to the impact of unknown surface emissivity. Overall, our case studies demonstrate that the proposed method achieves better prediction of nugget thickness and diameter than predicting without the transformation.

42 ENGINEERING↗
DOE OSTI2023

Application of unsupervised deep learning to image segmentation and in-situ contact angle measurements in a CO 2 -water-rock system

Rock surface wettability is a critical property that regulates multiphase flows in porous media, which can be quantified using the surface contact angle (CA). X-ray micro-computed tomography (μCT) provides an effective approach to in-situ measurements of surface CAs. However, the CA measurement accuracy depends significantly on the quality of CT image segmentation, which is the clustering of CT pixels into separate phases. Inspired by this, we developed a deep learning (DL)-based CA measurement workflow. Motivated by the recent tremendous progress in unsupervised learning techniques and aiming to avoid expensive manual data annotations, an unsupervised DL pipeline for CT image segmentation was proposed and implemented, which includes unsupervised model training and post-processing. The unsupervised model training was driven by a novel loss function constrained with feature similarity and spatial continuity and implemented by iterative forward and backward paths; the former clustered the pixel-wise feature vectors extracted by convolution neural networks, whereas the latter updated the parameters using gradient descent. An over-segmentation strategy was adopted for model training. The post-processing steps based on agglomerative hierarchical clustering (AHC) were implemented to further merge the over-segmented model output to the desired cluster number, which is intended to improve the efficiency of image segmentation. The developed unsupervised DL pipeline was compared with other commonly-used image segmentation methods using pixel-wise and physics-based evaluation metrics on a synthetic raw-image dataset, which had a known ground truth. The unsupervised DL pipeline showed the best performance. Next, the segmented images were input to an automatic CA measurement tool, and the results were validated by comparisons with manual measurements. The CA values from the manual and automatic measurements showed similar distributions and statistical properties. The automatic measurement demonstrated a wider spectrum because of the much larger number of measurement data points. The primary novelty of the unsupervised DL pipeline developed in this study lies in the novel loss function and the over-segmentation strategy associated with AHC post-processing. Finally, the workflow has been proven an efficient tool for pore-scale wettability characterization, which has a wide range of applications in fundamental studies of multiphase flows in natural porous media, which have critical implications to geological carbon sequestration, hydrocarbon energy recovery, and contaminant transport in groundwater.

42 ENGINEERING↗
DOE OSTI2023

Slicing Newton spheres with a two-camera 3D imaging system

We demonstrate a simple approach to achieve three-dimensional ion momentum imaging. The method employs two complementary metal–oxide–semiconductor cameras in addition to a standard microchannel plates/phosphor screen imaging detector. The two cameras are timed to measure the decay of luminescence excited by ion hits to extract the time of flight. The achieved time resolution is better than 10 ns, which is mainly limited by camera jitters. A better than 5 ns resolution can be achieved when the jitter is suppressed.

Chemistry↗
DOE OSTI2015

SU-E-J-09: Image Quality Comparison and Dose Quantification for 2.5 MV

Purpose: To compare the image quality of the 2.5MV imaging beam (2.5X-IMB) to that of a 6MV beam and to quantify the imaging dose of a 2.5X-IMB for constancy as specified by AAPM TG-142 Methods: The image quality of the 2.5X-IMB was compared to the 6MV imaging beam using the SNC ImagePro MV-QA phantom and the Varian supplied Las Vegas phantom (LVP). High resolution (1280×1280×16, 2 frames at 1.5MU/frame) and low resolution (640×640×16, 2 frames at 0.75MU/frame) images were compared for each phantom. MV-QA phantom images were evaluated quantitatively, and the LVP images were evaluated qualitatively. The imaging dose for 2.5X-IMB was quantified using the procedure outlined in TG51. PTWCC13-31013 chambers were used to measure a percent depth dose (PDD) curve for the 2.5X-IMB. All the factors described in TG51 were calculated using the 2.5X-IMB and a PTW30013 farmer chamber. Results: A comparison between 2.5X-IMB and 6MV image quality was performed both visually and with DoseLab software. The optimal window and level were set for each image of the LVP by the user. Visual inspection showed greater contrast resolution with the 2.5MV beam, but no significant difference with the change in imaging resolution. DoseLab reported similar spatial resolutions between the two energies, but the contrast-to-noise ratio (CNR) was greater for 2.5MV. The PDDx(10cm) for a 10x10cm2 field was measured to be 51.5%. Although this PDD value is off the scale of Figure 4 in TG51, the trend of the curve corresponding to the PTW31003 (equivalent) chamber led to an approximate kQ value of 1.00. Conclusion: When compared to 6MV imaging, 2.5X-IMB results in a better CNR. At low resolution, the DoseLab results for the two energies are comparable, but visual analysis favors the 2.5X-IMB images. Imaging dose was quantified for the 2.5X-IMB after following the TG51 methodology with appropriate approximations.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

TU-CD-BRA-05: Atlas Selection for Multi-Atlas-Based Image Segmentation Using Surrogate Modeling

Purpose: The growing size and heterogeneity in training atlas necessitates sophisticated schemes to identify only the most relevant atlases for the specific multi-atlas-based image segmentation problem. This study aims to develop a model to infer the inaccessible oracle geometric relevance metric from surrogate image similarity metrics, and based on such model, provide guidance to atlas selection in multi-atlas-based image segmentation. Methods: We relate the oracle geometric relevance metric in label space to the surrogate metric in image space, by a monotonically non-decreasing function with additive random perturbations. Subsequently, a surrogate’s ability to prognosticate the oracle order for atlas subset selection is quantified probabilistically. Finally, important insights and guidance are provided for the design of fusion set size, balancing the competing demands to include the most relevant atlases and to exclude the most irrelevant ones. A systematic solution is derived based on an optimization framework. Model verification and performance assessment is performed based on clinical prostate MR images. Results: The proposed surrogate model was exemplified by a linear map with normally distributed perturbation, and verified with several commonly-used surrogates, including MSD, NCC and (N)MI. The derived behaviors of different surrogates in atlas selection and their corresponding performance in ultimate label estimate were validated. The performance of NCC and (N)MI was similarly superior to MSD, with a 10% higher atlas selection probability and a segmentation performance increase in DSC by 0.10 with the first and third quartiles of (0.83, 0.89), compared to (0.81, 0.89). The derived optimal fusion set size, valued at 7/8/8/7 for MSD/NCC/MI/NMI, agreed well with the appropriate range [4, 9] from empirical observation. Conclusion: This work has developed an efficacious probabilistic model to characterize the image-based surrogate metric on atlas selection. Analytical insights lead to valid guiding principles on fusion set size design.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2016

TU-D-207B-01: A Prediction Model for Distinguishing Radiation Necrosis From Tumor Progression After Gamma Knife Radiosurgery Based On Radiomics Features From MR Images

Purpose: To develop and validate a prediction model using radiomics features extracted from MR images to distinguish radiation necrosis from tumor progression for brain metastases treated with Gamma knife radiosurgery. Methods: The images used to develop the model were T1 post-contrast MR scans from 71 patients who had had pathologic confirmation of necrosis or progression; 1 lesion was identified per patient (17 necrosis and 54 progression). Radiomics features were extracted from 2 images at 2 time points per patient, both obtained prior to resection. Each lesion was manually contoured on each image, and 282 radiomics features were calculated for each lesion. The correlation for each radiomics feature between two time points was calculated within each group to identify a subset of features with distinct values between two groups. The delta of this subset of radiomics features, characterizing changes from the earlier time to the later one, was included as a covariate to build a prediction model using support vector machines with a cubic polynomial kernel function. The model was evaluated with a 10-fold cross-validation. Results: Forty radiomics features were selected based on consistent correlation values of approximately 0 for the necrosis group and >0.2 for the progression group. In performing the 10-fold cross-validation, we narrowed this number down to 11 delta radiomics features for the model. This 11-delta-feature model showed an overall prediction accuracy of 83.1%, with a true positive rate of 58.8% in predicting necrosis and 90.7% for predicting tumor progression. The area under the curve for the prediction model was 0.79. Conclusion: These delta radiomics features extracted from MR scans showed potential for distinguishing radiation necrosis from tumor progression. This tool may be a useful, noninvasive means of determining the status of an enlarging lesion after radiosurgery, aiding decision-making regarding surgical resection versus conservative medical management.

60 APPLIED LIFE SCIENCES↗