Search NASA⌕ Search

SEARCH · Search NASA

Results for “FEATURE”

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 145 records · Page 8

SU-D-BRA-05: Toward Understanding the Robustness of Radiomics Features in CT

Purpose: To gauge the impact of inter-scanner variability on radiomics features in computed tomography (CT). Methods: We compared the radiomics features calculated for 17 scans of the specially designed Credence Cartridge Radiomics (CCR) phantom with those calculated for 20 scans of non–small cell lung cancer (NSCLC) tumors. The scans were acquired at four medical centers using General Electric, Philips, Siemens, and Toshiba CT scanners. Each center used its own routine thoracic imaging protocol. To produce a large dynamic range of radiomics feature values, the CCR phantom has 10 cartridges comprising different materials. The features studied were derived from the neighborhood gray-tone difference matrix or image intensity histogram. To quantify the significance of the inter-scanner variability, we introduced the metric “feature noise”, which compares the ratio of inter-scanner variability and inter-patient variability in decibels, positive values indicating substantial noise. We performed hierarchical clustering based to look for dependence of the features on the scan acquisition parameters. Results: For 5 of the 10 features studied, the inter-scanner variability was larger than the inter-patient variability. Of the 10 materials in the phantom, shredded rubber seemed to produce feature values most similar to those of the NSCLC tumors. The feature busyness had the greatest feature noise (14.3 dB), whereas texture strength had the least (−14.6 dB). Hierarchical clustering indicated that the features depended in part on the scanner manufacturer, image slice thickness, and pixel size. Conclusion: The variability in the values of radiomics features calculated for CT images of a radiomics phantom can be substantial relative to the variability in the values of these features calculated for CT images of NSCLC tumors. These inter-scanner differences and their effects should be carefully considered in future radiomics studies.

60 APPLIED LIFE SCIENCES↗

SU-F-R-30: Interscanner Variability of Radiomics Features in Computed Tomography (CT) Using a Standard ACR Phantom

Purpose: A simple approach to investigate Interscanner variability of Radiomics features in computed tomography (CT) using a standard ACR phantom. Methods: The standard ACR phantom was scanned on CT scanners from three different manufacturers. Scanning parameters of 120 KVp, 200 mA were used while slice thickness of 3.0 mm on two scanners and 3.27 mm on third scanner was used. Three spherical regions of interest (ROI) from water, medium density and high density inserts were contoured. Ninety four Radiomics features were extracted using an in-house program. These features include shape (11), intensity (22), GLCM (26), GLZSM (11), RLM (11), and NGTDM (5) and 8 fractal dimensions features. To evaluate the Interscanner variability across three scanners, a coefficient of variation (COV) is calculated for each feature group. Each group is further classified according to the COV- by calculating the percentage of features in each of the following categories: COV less than 2%, between 2 and 10% and greater than 10%. Results: For all feature groups, similar trend was observed for three different inserts. Shape features were the most robust for all scanners as expected. 70% of the shape features had COV <2%. For intensity feature group, 2% COV varied from 9 to 32% for three scanners. All features in four groups GLCM, GLZSM, RLM and NGTDM were found to have Interscanner variability ≥2%. The fractal dimensions dependence for medium and high density inserts were similar while it was different for water inserts. Conclusion: We concluded that even for similar scanning conditions, Interscanner variability across different scanners was significant. The texture features based on GLCM, GLZSM, RLM and NGTDM are highly scanner dependent. Since the inserts of the ACR Phantom are not heterogeneous in HU values suggests that matrix based 2nd order features are highly affected by variation in noise. Research partly funded by NIH/NCI R01CA190105-01.

60 APPLIED LIFE SCIENCES↗

MO-DE-207B-01: JACK FOWLER JUNIOR INVESTIGATOR COMPETITION WINNER: Between Somatic Mutations and PET-Based Radiomic Features in Non-Small Cell Lung Cancer

Purpose: Although PET-based radiomic features have been proposed to quantify tumor heterogeneity and shown promise in outcome prediction, little is known about their relationship with tumor genetics. This study assessed the association of [{sup 18}F]fluorodeoxyglucose (FDG)-PET-based radiomic features with non-small cell lung cancer (NSCLC) mutations. Methods: 348 NSCLC patients underwent FDG-PET/CT scans before treatment and were tested for genetic mutations. 13% (44/348) and 28% (96/348) patients were found to harbor EGFR (EGFR+) and KRAS (KRAS+) mutations, respectively. We evaluated nineteen PET-based radiomic features quantifying phenotypic traits, and compared them with conventional PET features (metabolic tumor volume (MTV) and maximum-SUV). The association between the feature values and mutation status was evaluated using the Wilcoxcon-rank-sum-test. The ability of each measure to predict mutations was assessed by the area under the receiver operating curve (AUC). Noether’s test was used to determine if the AUCs were significantly from random (AUC=0.50). All p-values were corrected for multiple testing by controlling the false discovery rate (FDR{sub Wilcoxon} and FDR{sub Noether}) of 10%. Results: Eight radiomic features, MTV, and maximum-SUV, were significantly associated with the EGFR mutation (FDR{sub Wilcoxon}=0.01–0.10). However, KRAS+ demonstrated no significantly distinctive imaging features compared to KRAS− (FDR{sub Wilcoxon}≥0.92). EGFR+ and EGFR− were significantly discriminated by conventional PET features (AUC=0.61, FDR{sub Noether}=0.04 for MTV and AUC=0.64, FDR{sub Noether}=0.01 for maximum-SUV). Eight radiomic features were significantly predictive for EGFR+ compared to EGFR− (AUC=0.59–0.67, FDR{sub Noether}=0.0032–0.09). Normalized-inverse-difference-moment outperformed all features in predicting EGFR mutation (AUC=0.67, FDR{sub Noether}=0.0032). Moreover, only the radiomic feature normalized-inverse-difference-moment could significantly predict KRAS+ from EGFR+ (AUC=0.65, FDR{sub Noether}=0.05). All measures failed to predict KRAS+ from KRAS− (AUC=0.50–0.54, FDR{sub Noether}≥0.92). Conclusion: PET imaging features were strongly associated with EGFR mutations in NSCLC. Radiomic features have great potential in predicting EGFR mutations. Our study may help develop a non-invasive imaging biomarker for EGFR mutation. R.M. has consulting interests with Amgen.

60 APPLIED LIFE SCIENCES↗

MO-DE-207B-04: Impact of Reconstruction Field of View On Radiomics Features in Computed Tomography (CT) Using a Texture Phantom

Purpose: To investigate the impact of reconstruction Field of View on Radiomics features in computed tomography (CT) using a texture phantom. Methods: A rectangular Credence Cartridge Radiomics (CCR) phantom, composed of 10 different cartridges, was scanned on four different CT scanners from two manufacturers. A pre-defined scanning protocol was adopted for consistency. The slice thickness and reconstruction interval of 1.5 mm was used on all scanners. The reconstruction FOV was varied to result a voxel size ranging from 0.38 to 0.98 mm. A spherical region of interest (ROI) was contoured on the shredded rubber cartridge from CCR phantom CT scans. Ninety three Radiomics features were extracted from ROI using an in-house program. These include 10 shape, 22 intensity, 26 GLCM, 11 GLZSM, 11 RLM, 5 NGTDM and 8 fractal dimensional features. To evaluate the Interscanner variability across three scanners, a coefficient of variation (COV) was calculated for each feature group. Each group was further classified according to the COV by calculating the percentage of features in each of the following categories: COV≤ 5%, between 5 and 10% and ≥ 10%. Results: Shape features were the most robust, as expected, because of the spherical contouring of ROI. Intensity features were the second most robust with 54.5 to 64% of features with COV < 5%. GLCM features ranged from 31 to 35% for the same category. RLM features were sensitive to specific scanner and 5% variability was 9 to 54%. Almost all GLZM and NGTDM features showed COV ≥10% among the scanners. The dependence of fractal dimensions features on FOV was not consistent across different scanners. Conclusion: We concluded that reconstruction FOV greatly influence Radiomics features. The GLZSM and NGTDM are highly sensitive to FOV. funded in part by Grant NIH/NCI R01CA190105-01.

60 APPLIED LIFE SCIENCES↗

The Origin of Faint Tidal Features around Galaxies in the RESOLVE Survey

We study tidal features around galaxies in the REsolved Spectroscopy Of a Local VolumE (RESOLVE) survey. Our sample consists of 1048 RESOLVE galaxies that overlap with the DECam Legacy Survey, which reaches an $r$-band 3$σ$ depth of ~27.9 mag arcsec -2 for a 100 arcsec 2 feature. Images were masked, smoothed, and inspected for tidal features such as streams, shells, or tails/arms. We find tidal features in 17± 2 % of our galaxies, setting a lower limit on the true frequency. The frequency of tidal features in the gas-poor (gas-to-stellar mass ratio <0.1) subsample is lower than in the gas-rich subsample (13± 3 % versus 19± 2 %). Within the gas-poor subsample, galaxies with tidal features have higher stellar and halo masses, ~3× closer distances to nearest neighbors (in the same group), and possibly fewer group members at fixed halo mass than galaxies without tidal features, but similar specific star formation rates. These results suggest tidal features in gas-poor galaxies are typically streams/shells from dry mergers or satellite disruption. In contrast, the presence of tidal features around gas-rich galaxies does not correlate with stellar or halo mass, suggesting these tidal features are often tails/arms from resonant interactions. Similar to tidal features in gas-poor galaxies, tidal features in gas-rich galaxies imply 1.7× closer nearest neighbors in the same group; however, they are associated with diskier morphologies, higher star formation rates, and higher gas content. In addition to interactions with known neighbors, we suggest that tidal features in gas-rich galaxies may arise from accretion of cosmic gas and/or gas-rich satellites below the survey limit.

79 ASTRONOMY AND ASTROPHYSICS↗

Feature Extraction Based on Decision Boundaries

In this paper, a novel approach to feature extraction for classification is proposed based directly on the decision boundaries. We note that feature extraction is equivalent to retaining informative features or eliminating redundant features; thus, the terms 'discriminantly information feature' and 'discriminantly redundant feature' are first defined relative to feature extraction for classification. Next, it is shown how discriminantly redundant features and discriminantly informative features are related to decision boundaries. A novel characteristic of the proposed method arises by noting that usually only a portion of the decision boundary is effective in discriminating between classes, and the concept of the effective decision boundary is therefore introduced. Next, a procedure to extract discriminantly informative features based on a decision boundary is proposed. The proposed feature extraction algorithm has several desirable properties: (1) It predicts the minimum number of features necessary to achieve the same classification accuracy as in the original space for a given pattern recognition problem; and (2) it finds the necessary feature vectors. The proposed algorithm does not deteriorate under the circumstances of equal class means or equal class covariances as some previous algorithms do. Experiments show that the performance of the proposed algorithm compares favorably with those of previous algorithms.

Lee, Chulhee↗

Process Sensitivity, Performance, and Direct Verification Testing of Adhesive Locking Features

Phase I: The use of adhesive locking features or liquid locking compounds (LLCs) (e.g., Loctite) as a means of providing a secondary locking feature has been used on NASA programs since the Apollo program. In many cases Loctite was used as a last resort when (a) self-locking fasteners were no longer functioning per their respective drawing specification, (b) access was limited for removal & replacement, or (c) replacement could not be accomplished without severe impact to schedule. Long-term use of Loctite became inevitable in cases where removal and replacement of worn hardware was not cost effective and Loctite was assumed to be fully cured and working. The NASA Engineering & Safety Center (NESC) and United Space Alliance (USA) recognized the need for more extensive testing of Loctite grades to better understand their capabilities and limitations as a secondary locking feature. These tests, identified as Phase I, were designed to identify processing sensitivities, to determine proper cure time, the correct primer to use on aerospace nutplate, insert and bolt materials such as A286 and MP35N, and the minimum amount of Loctite that is required to achieve optimum breakaway torque values. The .1900-32 was the fastener size tested, due to wide usage in the aerospace industry. Three different grades of Loctite were tested. Results indicate that, with proper controls, adhesive locking features can be successfully used in the repair of locking features and should be considered for design. Phase II: Threaded fastening systems used in aerospace programs typically have a requirement for a redundant locking feature. The primary locking method is the fastener preload and the traditional redundant locking feature is a self-locking mechanical device that may include deformed threads, non-metallic inserts, split beam features, or other methods that impede movement between threaded members. The self-locking resistance of traditional locking features can be directly verified during assembly by measuring the dynamic prevailing torque. Adhesive locking features or LLCs are another method of providing redundant locking, but a direct verification method has not been used in aerospace applications to verify proper installation when using LLCs because of concern for damage to the adhesive bond. The reliability of LLCs has also been questioned due to failures observed during testing with coupons for process verification, although the coupon failures have often been attributed to a lack of proper procedures. It is highly desirable to have a direct method of verifying the LLC cure or bond integrity. The purpose of the Phase II test program was to determine if the torque applied during direct verification of an adhesive locking feature degrades that locking feature. This report documents the test program used to investigate the viability of such a direct verification method. Results of the Phase II testing were positive, and additional investigation of direct verification of adhesive locking features is merited.

Golden, Johnny L.↗

Automatic Feature Tracking on Small Bodies for Autonomous Approach

Abstract—The autonomous approach of a spacecraft to an asteroid or comet (a small body) relies heavily on visual feature tracking to aid in estimating relative trajectories and the properties of the small body. Feature tracking for small bodies brings several challenges, including changing lighting, poor visual texture, and a concentration of features in a small part of an image. Six existing, open-source algorithms for feature tracking were tested on a simulated dataset and compared to the ground truth in the path of features. The main finding is that none of the algorithms provide all of the desired characteristics of long feature tracks with low errors and few outliers. Instead, there is a trade-off between long feature tracks and low error. The feature-matching algorithms SIFT, and BRISK provide good error characteristics, but short feature tracks, whereas the optical flow algorithm KLT provides long feature tracks, but with many features of large error. Given the challenges in feature tracking, it is recommended to focus development on each component of a feature tracking system: detection, description, and outlier rejection.

Morrell, Benjamin J↗

Co-Occurring Atmospheric Features and Their Contributions to Precipitation Extremes

Object-based identification algorithms for atmospheric features are commonly utilized to attribute global precipitation. This study employs a systematic approach to examine feature co-occurrences and their relationships to mean and extreme precipitation. Four features are identified using existing data sets for atmospheric rivers (ARs), mesoscale convective systems (MCSs), low-pressure systems (LPSs), and fronts (FTs). Often, a single atmospheric phenomenon satisfies the criteria set by multiple feature identification algorithms, yielding an association between precipitation and multiple features. Over the extra-tropics, the number of features attributed to a single event typically increases with precipitation intensity. Over two-thirds of the precipitation is from co-occurring features, with a considerable fraction related to AR-FT co-occurrences. Over the tropics, about one-quarter of precipitation is associated with co-occurring features, with LPS-MCS co-occurrences contributing substantially in monsoon regions. MCSs are the leading single-feature contributors over tropical land and oceans. In the extra-tropics, FTs, ARs, and their co-occurrences account for over half of the total precipitation over oceans. AR-FT-MCS and FT-MCS co-occurrences contribute to extremes (precipitation exceeding the 95th percentile) over both oceans (over 30%) and land (over 20%). Any combination of features involving MCSs shows a larger contribution to high percentiles of precipitation intensity. A case analysis indicates that AR-FT-MCS co-occurrences exhibit convective instability and deep vertical motion, suggesting that the feature trackers and reanalysis are capturing physics relevant to both convective and frontal systems. The results here emphasize the need for simultaneous identifications of multiple features when attributing precipitation to atmospheric phenomena.

54 ENVIRONMENTAL SCIENCES↗

PROBING THE '30 {mu}m' FEATURE: LESSONS FROM EXTREME CARBON STARS

Some carbon-rich evolved stars exhibit a very prominent spectral feature at {approx}{sup 3}0 {mu}m''. The C-rich nature of these objects suggests that the carrier is a sulfide, carbide, or other substance apt to form in a reducing environment. However, the carrier of this spectral feature remains disputed, with magnesium sulfide (MgS) as the most favored candidate. In order to investigate the carrier of the {sup 3}0 {mu}m'' feature further, we have taken a dual approach, studying both laboratory and observational data. In order to obtain a homogeneous sample, we studied the {sup 3}0 {mu}m'' feature observed in the spectra of galactic extreme carbon stars that exhibit the 11 {mu}m SiC absorption feature. Thus, we avoid potential differences in the target objects that could contribute to the observed differences in the shape, position, and strength of the {sup 3}0 {mu}m'' feature. In addition, we analyzed the shape and position of the {sup 3}0 {mu}m'' features for a range of sulfide minerals for which laboratory data exist. Our study of observed astronomical features shows a range of shapes and positions for the {sup 3}0 {mu}m'' feature despite similarities in the source objects. The nature of our sample argues against grain processing or temperature differences due to hardening of the stellar radiation field with evolution. While there are very few correlations between spectral parameters for our sample, the peak positions of the {approx}11 {mu}m absorption and 30 {mu}m emission features do correlate, and these peak positions correlate with the modeled optical depth. These correlations suggest that the carriers of the observable spectral features are closely related to one another and to the density/pressure in the dust formation zone. Furthermore, we suggest that the blue-broadening of the 11 {mu}m feature cannot be attributed to SiS{sub 2} based on existing laboratory spectra, but further lab work is needed to investigate the effect of increasing oxidation.

79 ASTRONOMY AND ASTROPHYSICS↗

Studies of the jet in BL Lacertae. I. Recollimation shock and moving emission features

Parsec-scale VLBA images of BL Lac at 15 GHz show that the jet contains a permanent quasi-stationary emission feature 0.26 mas (0.34 pc projected) from the core, along with numerous moving features. In projection, the tracks of the moving features cluster around an axis at a position angle of –166.°6 that connects the core with the standing feature. The moving features appear to emanate from the standing feature in a manner strikingly similar to the results of numerical two-dimensional relativistic magneto-hydrodynamic (RMHD) simulations in which moving shocks are generated at a recollimation shock (RCS). Because of this, and the close analogy to the jet feature HST-1 in M87, we identify the standing feature in BL Lac as an RCS. We assume that the magnetic field dominates the dynamics in the jet, and that the field is predominantly toroidal. From this we suggest that the moving features are compressions established by slow and fast mode magneto-acoustic MHD waves. We illustrate the situation with a simple model in which the slowest moving feature is a slow-mode wave, and the fastest feature is a fast-mode wave. In the model, the beam has Lorentz factor Γ{sub beam}{sup gal}≈3.5 in the frame of the host galaxy and the fast mode wave has Lorentz factor Γ{sub Fwave}{sup beam}≈1.6 in the frame of the beam. This gives a maximum apparent speed for the moving features, β{sub app} = v{sub app}/c = 10. In this model the Lorentz factor of the pattern in the galaxy frame is approximately three times larger than that of the beam itself.

79 ASTRONOMY AND ASTROPHYSICS↗

TU-CD-BRB-04: Automated Radiomic Features Complement the Prognostic Value of VASARI in the TCGA-GBM Dataset

Purpose: To compare the complementary prognostic value of automated Radiomic features to that of radiologist-annotated VASARI features in TCGA-GBM MRI dataset. Methods: For 96 GBM patients, pre-operative MRI images were obtained from The Cancer Imaging Archive. The abnormal tumor bulks were manually defined on post-contrast T1w images. The contrast-enhancing and necrotic regions were segmented using FAST. From these sub-volumes and the total abnormal tumor bulk, a set of Radiomic features quantifying phenotypic differences based on the tumor intensity, shape and texture, were extracted from the post-contrast T1w images. Minimum-redundancy-maximum-relevance (MRMR) was used to identify the most informative Radiomic, VASARI and combined Radiomic-VASARI features in 70% of the dataset (training-set). Multivariate Cox-proportional hazards models were evaluated in 30% of the dataset (validation-set) using the C-index for OS. A bootstrap procedure was used to assess significance while comparing the C-Indices of the different models. Results: Overall, the Radiomic features showed a moderate correlation with the radiologist-annotated VASARI features (r = −0.37 – 0.49); however that correlation was stronger for the Tumor Diameter and Proportion of Necrosis VASARI features (r = −0.71 – 0.69). After MRMR feature selection, the best-performing Radiomic, VASARI, and Radiomic-VASARI Cox-PH models showed a validation C-index of 0.56 (p = NS), 0.58 (p = NS) and 0.65 (p = 0.01), respectively. The combined Radiomic-VASARI model C-index was significantly higher than that obtained from either the Radiomic or VASARI model alone (p = <0.001). Conclusion: Quantitative volumetric and textural Radiomic features complement the qualitative and semi-quantitative annotated VASARI feature set. The prognostic value of informative qualitative VASARI features such as Eloquent Brain and Multifocality is increased with the addition of quantitative volumetric and textural features from the contrast-enhancing and necrotic tumor regions. These results should be further evaluated in larger validation cohorts.

60 APPLIED LIFE SCIENCES↗

SU-F-R-35: Repeatability of Texture Features in T1- and T2-Weighted MR Images

Purpose: To evaluate repeatability of lung tumor texture features from inspiration/expiration MR image pairs for potential use in patient specific care models and applications. Repeatability is a desirable and necessary characteristic of features included in such models. Methods: T1-weighted Volumetric Interpolation Breath-Hold Examination (VIBE) and/or T2-weighted MRI scans were acquired for 15 patients with non-small cell lung cancer before and during radiotherapy for a total of 32 and 34 same session inspiration-expiration breath-hold image pairs respectively. Bias correction was applied to the VIBE (VIBE-BC) and T2-weighted (T2-BC) images. Fifty-nine texture features at five wavelet decomposition ratios were extracted from the delineated primary tumor including: histogram(HIST), gray level co-occurrence matrix(GLCM), gray level run length matrix(GLRLM), gray level size zone matrix(GLSZM), and neighborhood gray tone different matrix (NGTDM) based features. Repeatability of the texture features for VIBE, VIBE-BC, T2-weighted, and T2-BC image pairs was evaluated by the concordance correlation coefficient (CCC) between corresponding image pairs, with a value greater than 0.90 indicating repeatability. Results: For the VIBE image pairs, the percentage of repeatable texture features by wavelet ratio was between 20% and 24% of the 59 extracted features; the T2-weighted image pairs exhibited repeatability in the range of 44–49%. The percentage dropped to 10–20% for the VIBE-BC images, and 12–14% for the T2-BC images. In addition, five texture features were found to be repeatable in all four image sets including two GLRLM, two GLZSM, and one NGTDN features. No single texture feature category was repeatable among all three image types; however, certain categories performed more consistently on a per image type basis. Conclusion: We identified repeatable texture features on T1- and T2-weighted MRI scans. These texture features should be further investigated for use in specific applications such as tissue classification and changes during radiation therapy utilizing a standard imaging protocol. Authors have the following disclosures: a research agreement with Philips Medical systems (Hugo, Weiss), a license agreement with Varian Medical Systems (Hugo, Weiss), research grants from the National Institute of Health (Hugo, Weiss), UpToDate royalties (Weiss), and none(Mahon, Ford, Karki). Authors have no potential conflicts of interest to disclose.

60 APPLIED LIFE SCIENCES↗

Feature Selection for Classification of Polar Regions Using a Fuzzy Expert System

Labeling, feature selection, and the choice of classifier are critical elements for classification of scenes and for image understanding. This study examines several methods for feature selection in polar regions, including the list, of a fuzzy logic-based expert system for further refinement of a set of selected features. Six Advanced Very High Resolution Radiometer (AVHRR) Local Area Coverage (LAC) arctic scenes are classified into nine classes: water, snow / ice, ice cloud, land, thin stratus, stratus over water, cumulus over water, textured snow over water, and snow-covered mountains. Sixty-seven spectral and textural features are computed and analyzed by the feature selection algorithms. The divergence, histogram analysis, and discriminant analysis approaches are intercompared for their effectiveness in feature selection. The fuzzy expert system method is used not only to determine the effectiveness of each approach in classifying polar scenes, but also to further reduce the features into a more optimal set. For each selection method,features are ranked from best to worst, and the best half of the features are selected. Then, rules using these selected features are defined. The results of running the fuzzy expert system with these rules show that the divergence method produces the best set features, not only does it produce the highest classification accuracy, but also it has the lowest computation requirements. A reduction of the set of features produced by the divergence method using the fuzzy expert system results in an overall classification accuracy of over 95 %. However, this increase of accuracy has a high computation cost.

Penaloza, Mauel A.↗

LI-RADS for CT diagnosis of hepatocellular carcinoma: performance of major and ancillary features

Purpose: To evaluate the diagnostic performance of Liver Imaging Reporting and Data System (LI-RADS) v2017 major features, the impact of ancillary features, and categories on contrast-enhanced computed tomography (CECT) for the diagnosis of hepatocellular carcinoma (HCC). Materials and methods: This retrospective study included 59 patients (104 observations including 72 HCCs) with clinical suspicion of HCC undergoing CECT between 2013 and 2016. Two radiologists independently assessed major and ancillary imaging features for each liver observation and assigned a LI-RADS category based on major features only and in combination with ancillary features. The composite reference standard included pathology or imaging. Per-lesion estimates of diagnostic performance of major features, ancillary features, and LI-RADS categories were assessed by generalized estimating equation models. Results: Major features (arterial phase hyperenhancement, washout, capsule, and threshold growth) respectively had a sensitivity of 86.1%, 81.6%, 20.7%, and 26.1% and specificity of 39.3%, 67.9%, 89.9%, and 85.0% for HCC. Ancillary features (ultrasound visibility as discrete nodule, subthreshold growth, and fat in mass more than adjacent liver) respectively had a sensitivity of 42.6%, 50.8%, and 15.1% and a specificity of 79.2%, 66.9%, and 96.4% for HCC. Ancillary features modified the final category in 4 of 104 observations. For HCC diagnosis, categories LR-3, LR-4, LR-5, and LR-TIV (tumor in vein) had a sensitivity of 5.3%, 29.0%, 53.7%, and 10.7%; and a specificity of 49.1%, 84.4%, 97.3%, and 96.4%, respectively. Conclusion: On CT, LR-5 category has near-perfect specificity for the diagnosis of HCC and ancillary features modifies the final category in few observations.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Modeling Pathologic Response of Esophageal Cancer to Chemoradiation Therapy Using Spatial-Temporal {sup 18}F-FDG PET Features, Clinical Parameters, and Demographics

Purpose: To construct predictive models using comprehensive tumor features for the evaluation of tumor response to neoadjuvant chemoradiation therapy (CRT) in patients with esophageal cancer. Methods and Materials: This study included 20 patients who underwent trimodality therapy (CRT + surgery) and underwent {sup 18}F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) both before and after CRT. Four groups of tumor features were examined: (1) conventional PET/CT response measures (eg, standardized uptake value [SUV]{sub max}, tumor diameter); (2) clinical parameters (eg, TNM stage, histology) and demographics; (3) spatial-temporal PET features, which characterize tumor SUV intensity distribution, spatial patterns, geometry, and associated changes resulting from CRT; and (4) all features combined. An optimal feature set was identified with recursive feature selection and cross-validations. Support vector machine (SVM) and logistic regression (LR) models were constructed for prediction of pathologic tumor response to CRT, cross-validations being used to avoid model overfitting. Prediction accuracy was assessed by area under the receiver operating characteristic curve (AUC), and precision was evaluated by confidence intervals (CIs) of AUC. Results: When applied to the 4 groups of tumor features, the LR model achieved AUCs (95% CI) of 0.57 (0.10), 0.73 (0.07), 0.90 (0.06), and 0.90 (0.06). The SVM model achieved AUCs (95% CI) of 0.56 (0.07), 0.60 (0.06), 0.94 (0.02), and 1.00 (no misclassifications). With the use of spatial-temporal PET features combined with conventional PET/CT measures and clinical parameters, the SVM model achieved very high accuracy (AUC 1.00) and precision (no misclassifications)—results that were significantly better than when conventional PET/CT measures or clinical parameters and demographics alone were used. For groups with many tumor features (groups 3 and 4), the SVM model achieved significantly higher accuracy than did the LR model. Conclusions: The SVM model that used all features including spatial-temporal PET features accurately and precisely predicted pathologic tumor response to CRT in esophageal cancer.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Speech Emotion Feature Selection Method Based on Contribution Analysis Algorithm of Neural Network

There are many emotion features. If all these features are employed to recognize emotions, redundant features may be existed. Furthermore, recognition result is unsatisfying and the cost of feature extraction is high. In this paper, a method to select speech emotion features based on contribution analysis algorithm of NN is presented. The emotion features are selected by using contribution analysis algorithm of NN from the 95 extracted features. Cluster analysis is applied to analyze the effectiveness for the features selected, and the time of feature extraction is evaluated. Finally, 24 emotion features selected are used to recognize six speech emotions. The experiments show that this method can improve the recognition rate and the time of feature extraction.

97 MATHEMATICS AND COMPUTING↗

Poloidal asymmetry in the narrow heat flux feature in the TCV scrape-off layer

Heat flux profiles inferred from a reciprocating probe at the outer midplane of the TCV tokamak during inner wall limited discharges feature radial fall-off lengths that shorten near the last closed flux surface (LCFS) consistent with the so-called narrow feature. The narrow feature is significantly wider on the outboard side compared with that measured on the inner wall by infrared thermography, so it is difficult to discern from the main scrape-off layer feature. After small shifts were applied for alignment, the fraction of the power contained in the narrow feature matches between inboard and outboard measurements, and they scale together with plasma current Ip, suggesting that we are observing the same phenomenon. The outboard side fall-off length within the narrow feature is found to scale closely with the radial correlation length of the edge turbulence as expected if the narrow feature arises due to radially sheared E x B flows. This is found to hold true even for cases where the narrow feature is weak and the fall-off lengths are approaching that of the far scrape-off layer. After the small shifts for alignment, non-zero floating potential profiles were found to match between inboard and outboard sides. A simple model of polarization and diamagnetic cross-field currents is described, which is consistent with the shape of these floating potential profiles. The model predicts that the floating potential at the LCFS must be negative, which supports the argument to shift the upstream measurements. The predicted currents are also consistent with the E x B flows believed to cause the narrow feature. The model is used to predict the magnitude of the floating potential of the LCFS, and the results are found to match measurements for all values of I p . This paper therefore demonstrates consistency between the measurements of the narrow feature on the inboard and outboard sides of the plasma, as well as consistency between the measurements, non-linear turbulence simulations, and analytical models of the narrow feature arising from sheared E x B flows.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗