Search NASA⌕ Search

SEARCH · Search NASA

Results for “Variance analysis”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

Study on Corrosion Behavior of 7075 Aluminum Alloy with Retrogression and Reaging Using Taguchi Method

The corrosion resistance and mechanical properties of 7075 aluminum alloy are improved by retrogression and reaging (RRA) heat treatment. Due to the complex process of RRA heat treatment, a Taguchi L25 orthogonal array is designed to optimize the process parameters of RRA heat treatment. In this work, analysis of variance was performed to find out the significant heat treatment parameters. Microstructure, mechanical properties, intergranular corrosion, exfoliation corrosion and stress corrosion cracking behaviors of 7075 aluminum alloy after RRA heat treatment were investigated by tensile test, slow strain rate test, transmission electron microscope and scanning electron microscope. Results showed that the RRA treatment involving preaging at 120 °C for 16 h, retrogression at 200 °C for 8 min and reaging at 120 °C for 24 h resulted in a desired combination of corrosion resistance and mechanical properties.

36 MATERIALS SCIENCE↗
DOE OSTI2019

Optimization of Tensile and Corrosion Properties of Dissimilar Friction Stir Welded AA2024-7075 Joints

In this work, we performed some joining experiments of dissimilar aluminum alloys AA2024-T351 and AA7075-T651 using friction stir welding (FSW) technique with different process parameters (rotational speed, welding speed and plunge depth). Response surface methodology based on a central composite rotatable design was used to establish the mathematical models predicting the tensile properties and corrosion rate of dissimilar FSW joints with some high confidence level. Analysis of variance method was employed to verify the developed models. Besides, the influences of FSW process parameters on tensile and corrosion properties of joints were analyzed. In order to obtain a desirable combination property of the dissimilar FSW joints, the optimum process parameters after experimental verification were proposed as: rotational speed 1495 rev/min, welding speed 187 mm/min and plunge depth 0.03 mm. This verification illustrated that the developed models are appropriate for the modeling and optimization of process.

36 MATERIALS SCIENCE↗
DOE OSTI2014

Impact of process conditions on the density and durability of wheat, oat, canola, and barley straw briquettes

The present study is to understand the impact of process conditions on the quality attributes of wheat oat, barley, and canola straw briquettes. Analysis of variance indicated that briquette moisture content and initial density immediately after compaction and final density after 2 weeks of storage are strong functions of feedstock moisture content and compression pressure, whereas durability rating is influenced by die temperature and feedstock moisture content. Briquettes produced at a low feedstock moisture content of 9 % (w.b.) yielded maximum densities >700 kg/m3 for wheat, oat, canola, and barley straws. Lower feedstock moisture content of <10 % (w.b.) and higher die temperatures >110 °C and compression pressure >10 MPa minimized the briquette moisture content and maximized densities and durability rating based on surface plots observations. Optimal process conditions indicated that a low feedstock moisture content of about 9 % (w.b.), high die temperature of 120–130 °C, medium-to-large hammer mill screen sizes of about 24 to 31.75 mm, and low to high compression pressures of 7.5 to 12.5 MPa minimized briquette moisture content to <8 % (w.b.) and maximized density to >700 kg/m3. Durability rating >90 % is achievable at higher die temperatures of >123 °C, lower to medium feedstock moisture contents of 9 to 12 % (w.b.), low to high compression pressures of 7.5 to 12.5 MPa, and large hammer mill screen size of 31.75 mm, except for canola where a lower compression pressure of 7.5 to 8.5 MPa and a smaller hammer mill screen size of 19 mm for oat maximized the durability rating values.

09 BIOMASS FUELS↗
DOE OSTI2021

Impact of cyberattacks on safety and stability of connected and automated vehicle platoons under lane changes

Connected and automated vehicles (CAVs) offer a huge potential to improve the operations and safety of transportation systems. However, the use of smart devices and communications in CAVs introduce new risks. CAVs would leverage vehicle to vehicle (V2V) and vehicle to infrastructure (V2I) communication, thus providing additional system access points compared to traditional systems. Automation makes these systems more vulnerable and increases the consequences of cyberattacks. This study utilizes an infrastructure-based communication platform consisting of cooperative adaptive cruise control and lane control advisories developed by the authors to perform cyber risk assessment of CAVs. The study emulates three types of cyberattacks (message falsification, dedicated denial of service, and spoofing attacks) in a representative traffic environment consisting of multiple CAV platoons and lane change events to analyze the safety and stability impacts of the cyberattacks. Simulation experiments using VISSIM reveals that traffic stream and CAV string is unstable under all three types of cyberattacks. The worst case is represented by the message falsification attack. Increases in volatility are observed over a no attack case, with variations increasing by an average of 43%–51% along with an increase of over 3000 crash conflicts. Similarly, lane change crash conflicts are observed to be more severe compared to rear end crash conflicts, showing a higher probability of severe injuries. Further, the case of slight cyberattack on a single CAV also creates significant disruption in the traffic stream. Analysis of variance (ANOVA) reveals the statistical significance of the results. Furthermore, these results pave the way for future design of secure systems from a monitoring perspective.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Evolution of AISI 304L stainless steel part properties due to powder recycling in laser powder-bed fusion

Selective laser melting (SLM) is a powder-bed fusion process in Additive Manufacturing (AM) that uses a laser beam to selectively fuse layers of powder into near net-shape components with little porosity. However, inconsistencies in the part properties due to the presence of defects in as-built components has hindered the widespread adoption of SLM for industrial applications forcing researchers to study the sources of variation for quality control purposes. A critical area suspected of creating variation in the part properties is the feedstock, where batch-to-batch differences as well as changes in the powder properties with reuse have the potential to affect performance. During processing, laser spatter and condensate eject from the melt pool and deposit into the powder-bed surrounding the parts. These particulates, more commonly known as ejecta, differ morphologically and chemically from the virgin powder compromising its reusability. In this study, 304L stainless steel powder was recycled for a total of 5 times through a systematic approach aimed at accelerating powder degradation to reveal the influence on both the tensile and impact toughness properties. Through analysis of variance (ANOVA), it was found that tensile properties did not change with reuse, while the impact toughness showed a steady decline illustrating the differences in static and dynamic part properties due to powder recycling.

36 MATERIALS SCIENCE↗
DOE OSTI2019

Mitigation of variable seasonal productivity in algae biomass through blending and ensiling: An assessment of compositional changes in storage

Microalgae biomass has gained considerable attention in recent years as an attractive feedstock for biofuel production. However, seasonal variability in algae growth presents a challenge to delivering a consistent feedstock supply to a conversion facility. Drying is one approach to preserving algae produced in productive summer months for conversion in winter, but is energy intensive and costly. Ensiling is an alternative to drying that preserves algae in wet anaerobic conditions in the presence of organic acids. We have previously demonstrated that dry matter loss of algae:corn stover blends can be limited to 8% over 35 days using this approach. However, longer storage durations will be required to maintain a consistent feedstock supply to a biorefinery. Furthermore, the impacts to biomass quality that occur during storage are unknown. A Box-Behnken design of experiment was conducted to determine the influence of moisture, soluble sugar, and algae content on dry matter loss of algae:corn stover blends in wet anaerobic storage over a 30-day period. Using one-way analysis of variance (ANOVA) moisture and soluble sugar content were found to significantly affect dry matter loss in storage. The results of experimental design were then used to inform the initiation of a 180 day storage study that evaluated the storage performance and biomass composition of algae:corn stover blends containing 5, 20 or 40% algae. After 180 days of storage dry matter loss ranged from 6 – 17%. A relative increase in nitrogen content of stored blends indicates that protein content was unaffected by storage, while the carbohydrate fraction was most impacted, decreasing by 5 – 15%. For each stored blend the oxygen content decreased in storage resulting increased higher heating values (HHV). As a result, the impact that changes to biochemical and elemental composition of algae:corn stover blends has on biomass conversion to fuels and chemicals is discussed.

09 BIOMASS FUELS↗
DOE OSTI2021

Quantifying physical parameterization uncertainties associated with land-atmosphere interactions in the WRF model over Amazon

The Weather Research and Forecasting (WRF) model can be used to diagnose regional land-atmosphere (L-A) coupling strength in the absence of sufficient observations but subjected to uncertainties associated with model physical parameterizations. In this study, we propose a framework to quantify and reduce model physical parameterization uncertainties associated with surface fluxes and L-A coupling. An ensemble of WRF simulations with different physical schemes is used to simulate surface fluxes and land-atmosphere coupling strength over the Amazon region. The physical parameterizations investigated include cloud microphysics (MP), land surface processes (LSM), planetary boundary layer (PBL), surface layer (SL), and cumulus (CU). We perform 120 ensemble simulations using the WRF model and different combinations of six MPs, three LSMs, six PBLs and SLs and three CUs. The measurements from the GoAMAZON field campaign and satellite data are used to evaluate model performance. A Multi-way analysis of variance (ANOVA) approach is applied to quantify the relative importance of different physics processes on L-A coupling. The Tukey’s test is used to sort schemes that have no significant differences into one group. The suite of physics that result in the best simulations of the corresponding variables are selected based on the Taylor skill score. Results show that the relative importance of processes and their interaction vary with the variables of interest. For example, CU was the most important process in modulating soil moisture, 2 m-humidity, latent heat, and net radiation. LSM showed dominant effects on 2 m-temperature and also has the largest impact on sensible heat and the lifting condensation level. The best physical parameterization ensembles show much narrower ranges of the variables of interest than the priori ensemble. Results of this study show the roles of different physical processes in modulating L-A interactions, quantify model uncertainties from physical processes, and provide insights for improving the model physics parameterizations.

Wang, Chen↗
DOE OSTI2014

Effect of process variables on the density and durability of the pellets made from high moisture corn stover

A flat die pellet mill was used to understand the effect of high levels of feedstock moisture content in the range of 28–38% (w.b.), with die rotational speeds of 40–60 Hz, and preheating temperatures of 30–110 °C on the pelleting characteristics of 4.8 mm screen size ground corn stover using an 8 mm pellet die. The physical properties of the pelletised biomass studied are: (a) pellet moisture content, (b) unit, bulk and tapped density, and (c) durability. Pelletisation experiments were conducted based on central composite design. Analysis of variance (ANOVA) indicated that feedstock moisture content influenced all of the physical properties at P < 0.001. Pellet moisture content decreased with increase in preheating temperature to about 110 °C and decreasing the feedstock moisture content to about 28% (w.b.). Response surface models developed for quality attributes with respect to process variables has adequately described the process with coefficient of determination (R2) values of >0.88. The other pellet quality attributes such as unit, bulk, tapped density, were maximised at feedstock moisture content of 30–33% (w.b.), die speeds of >50 Hz and preheating temperature of >90 °C. In case of durability a medium moisture content of 33–34% (w.b.) and preheating temperatures of >70 °C and higher die speeds >50 Hz resulted in high durable pellets. It can be concluded from the present study that feedstock moisture content, followed by preheating, and die rotational speed are the interacting process variables influencing pellet moisture content, unit, bulk and tapped density and durability.

09 BIOMASS FUELS↗
DOE OSTI2022

Microbial community diversity changes during voltage reversal repair in a 12-unit microbial fuel cell

Microbial fuel cell stacks (MFC-Stack) are often confronted with voltage reversals, likely due to an interplay between microbial community dynamics and insufficient electric circuit balancing. Herein, we provide new insight into voltage reversals by examining the microbiomes of twelve MFC units of a 12-liter Pilot-MFC-Stack during repair. Different biofilm repair methods (self-healing, electrostimulation, and re-acclimatization upon cross-inoculation) were used to evaluate the microbial community response. In addition, MFC-Stack simulation was performed based on Kirchhoff’s Second Law to predict values for source potentials and post-evaluate internal resistances. Analysis of the 16S rRNA amplicon sequencing data suggests that the biofilm repair methods could slowly heal damaged biofilms. Notably, severely voltage reversed MFC units had low electrogen relative abundances (18%) and positive anode potentials, while strong bioanodes and contained more than 50% electrogens and had negative anode potentials. Between-community analyses (beta diversity ordination and multinomial regression) of the voltage reversed MFC units revealed differences among biofilms in contrast to healthy/strong MFC units. Permutational multivariate analysis of variance (PERMANOVA) confirmed that reversed biofilms were, indeed, significantly (p < 0.05) different from stronger ones. Overall, these analyses demonstrated the utility of combining electrotechnical and microbial community analyses, especially beta diversity ordination and multinomial regression, to understand problematic MFC units and the potential success of a biofilm repair method. Finally, thicker biofilms were usually healthier and stronger, although thickness was no guarantee for proper structure and power function as all factors were interdependent. There was an evolutionary trend that strong anodes became stronger/healthier and others weaker. This spontaneous trend has to be considered to avoid irreversible voltage reversals and to repair electrogenic biofilms in an MFC-Stack.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2016

Occurrence of vancomycin-resistant and -susceptible Enterococcus spp. in reclaimed water used for spray irrigation

Reclaiming municipal wastewater for agricultural, environmental, and industrial purposes is increasing in the United States to combat dwindling freshwater supplies. However, there is a lack of data regarding the microbial quality of reclaimed water. In particular, no previous studies have evaluated the occurrence of vancomycin-resistant enterococci (VRE) in reclaimed water used at spray irrigation sites in the United States. To address this knowledge gap, we investigated the occurrence, concentration, and antimicrobial resistance patterns of VRE and vancomycin-susceptible enterococci at three U.S. spray irrigation sites that use reclaimed water. We collected 48 reclaimed water samples from one Mid-Atlantic and two Midwest spray irrigation sites, as well as their respective wastewater treatment plants, in 2009 and 2010. Samples were analyzed for total enterococci and VRE using standard membrane filtration. Isolates were purified and then confirmed using biochemical tests and PCR. Antimicrobial susceptibility testing was conducted using the Sensititre® microbroth dilution system. Data were analyzed by two-sample proportion tests and one-way analysis of variance. We detected total enterococci and VRE in 71% (34/48) and 4% (2/48) of reclaimed water samples, respectively. Enterococcus faecalis was the most common species identified. At the Mid-Atlantic spray irrigation site, UV radiation decreased total enterococci to undetectable levels; however, subsequent storage in an open-air pond at this site resulted in increased concentrations of enterococci. E. faecalis isolates recovered from the Mid-Atlantic spray irrigation site expressed intrinsic resistance to quinupristin/dalfopristin; however, non-E. faecalis isolates expressed resistance to quinupristin/dalfopristin (52% of isolates), vancomycin (4%), tetracycline (13%), penicillin (4%) and ciprofloxacin (17%). Our findings show that VRE are present in low numbers in reclaimed water at point-of-use at the sampled spray irrigation sites; however, resistance to other antimicrobial classes is more prevalent, particularly among non-E. faecalis isolates. - Highlights: • Enterococci were recovered in 71% of reclaimed water samples. • Vancomycin-resistant enterococci were detected in 4% of reclaimed water samples. • UV radiation at irrigation sites reduced enterococci to undetectable levels. • Storage of reclaimed water in open-air ponds increased levels of enterococci.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2015

Transient thermal protection of film covering circular aperture by sublimation and weak decomposition

Unwanted heating of sensitive surfaces in harsh thermal environments can be prevented by precise application of sacrificial materials such as sublimation layers and pyrolyzing films. The use of sublimation for the protection of circular polyimide membranes subjected to brief (~100 ms) heating by infrared radiation and hot (6000 K) inert gas convection is analyzed. Selection of sublimation material and sublimation layer and membrane thickness is considered with emphasis on providing sufficient thermal protection yet negligible unwanted material remaining at the end of a specified heating period. Though the analysis here is general, the motivation is protection of the polyimide films covering the laser entrance holes on IFE (inertial fusion energy) hohlraums being injected into the hot gas (xenon) protecting IFE reactor chambers. Furthermore, both one and two dimensional thermal models are used to develop a robust thermal concept. Sensitivity analyses (SA) methods are exercised to show where the design may be vulnerable and which input parameters have the greatest effect on performance and likelihood of success. For the design and conditions considered, methane sublimating layers are probably preferred over xenon or pentane.

42 ENGINEERING↗
DOE OSTI2013

Comparison of Various Online Strategies to Account for Interfractional Variations for Pancreatic Cancer

Purpose: To identify practical techniques to address the large interfractional variations for pancreas irradiation by comparing various used/proposed online strategies. Methods and Materials: The daily computed tomography (CT) images acquired using a respiration-gated in-room CT (CTVision; Siemens) for 10 pancreatic cancer patients treated with image guided radiation therapy (IGRT) were analyzed. The contours of the pancreas and organs at risk on each daily CT set were generated by populating from the planning CT using a deformable registration tool (ABAS; Elekta) with manual editing. Nine online strategies were considered: (1) standard IGRT (ie, IGRT with 0-mm additional margin [AM]); (2) IGRT with 2-mm AM; (3) IGRT with 5-mm AM; (4) IGRT with plan renormalized to maintain 95% planning target volume (PTV) coverage; (5) full-scale reoptimization; (6) reoptimization starting from the original plan; (7) segment aperture morphing (SAM) from the original plan, based on PTV shape change; (8) SAM plus segment weight optimization; and (9) reoptimization starting from the SAM plan. One-way analysis of variance was applied to plan qualities for the 9 strategies to assess statistical significance in difference. Results: The 3 IGRT strategies (1-3) lead to either inadequate PTV coverage or higher doses to critical structures, indicating that the additional margins alone are not adequate to account for the changes. The full-scale reoptimization results in the best plan but requires the delineation of several structures, which is time consuming. The SAM strategy (7) was the fastest one, because it requires delineating only 1 structure (target), and its plan quality was comparable to that for the full-scale reoptimization. Conclusion: Online replanning strategies can lead to either reduced organs-at-risk dose and/or improved target coverage as compared with the current practice of IGRT. The SAM-based online replanning is comparable to full-scale reoptimization and is efficient for practical use.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2013

Doses to Carotid Arteries After Modern Radiation Therapy for Hodgkin Lymphoma: Is Stroke Still a Late Effect of Treatment?

Purpose: Hodgkin lymphoma (HL) survivors are at an increased risk of stroke because of carotid artery irradiation. However, for early-stage HL involved node radiation therapy (INRT) reduces the volume of normal tissue exposed to high doses. Here, we evaluate 3-dimensional conformal radiation therapy (3D-CRT), volumetric-modulated arc therapy (VMAT), and proton therapy (PT) delivered as INRT along with the extensive mantle field (MF) by comparing doses to the carotid arteries and corresponding risk estimates. Methods and Materials: We included a cohort of 46 supradiaphragmatic stage I-II classical HL patients. All patients were initially treated with chemotherapy and INRT delivered as 3D-CRT (30 Gy). For each patient, we simulated MF (36 Gy) and INRT plans using VMAT and PT (30 Gy). Linear dose-response curves for the 20-, 25-, and 30-year risk of stroke were derived from published HL data. Risks of stroke with each technique were calculated for all patients. Statistical analyses were performed with repeated measures analysis of variance. Results: The mean doses to the right and left common carotid artery were significantly lower with modern treatment compared with MF, with substantial patient variability. The estimated excess risk of stroke after 20, 25, and 30 years was 0.6%, 0.86%, and 1.3% for 3D-CRT; 0.67%, 0.96%, and 1.47% for VMAT; 0.61%, 0.96%, and 1.33% for PT; and 1.3%, 1.72%, and 2.61% for MF. Conclusions: INRT reduces the dose delivered to the carotid arteries and corresponding estimated risk of stroke for HL survivors. Even for the subset of patients with lymphoma close to the carotid arteries, the estimated risk is low.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2014

Efficacy of Synbiotics to Reduce Acute Radiation Proctitis Symptoms and Improve Quality of Life: A Randomized, Double-Blind, Placebo-Controlled Pilot Trial

Purpose: To evaluate whether the daily intake of synbiotics interferes in radiation-induced acute proctitis symptoms and in quality of life in patients with prostate cancer. Methods and Materials: Twenty patients who underwent 3-dimensional conformal radiation therapy for prostate cancer were randomized to intake either a synbiotic powder containing Lactobacillus reuteri 10{sup 8} colony-forming units and 4.3 g of soluble fiber (Nestlé) or placebo. The questionnaire EORTC QLQ-PRT23 was applied before the beginning of radiation therapy and in every week for the first 4 weeks of treatment. The sum of both the complete (proctitis symptoms plus quality of life) and partial (proctitis symptoms) scores of the EORTC QLQ-PRT23 (European Organization for Research and Treatment of Cancer Quality of Life Module for Proctitis–23 items) questionnaire were the main endpoints. Results: This pilot study showed that the complete questionnaire score (median [range]) was higher in the second (23 [21-30] vs 26.5 [22-34], P<.05) and third (23 [21-32] vs 27.5 [24-33], P<.01) weeks in the placebo group. Proctitis symptoms were highest scored in the placebo group in both the second (19.5 [16-25]) and third (19 [17-24]) weeks than in the synbiotic group (week 2: 16.5 [15-20], P<.05; week 3: 17 [15-23], P<.01). In both scores the placebo group had a significantly higher result (P<.01) than the synbiotic group (repeated-measures analysis of variance). Conclusions: Synbiotics reduce proctitis symptoms and improve quality of life in radiation-induced acute proctitis during radiation therapy for prostate cancer.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2014

Deformable Image Registration for Adaptive Radiation Therapy of Head and Neck Cancer: Accuracy and Precision in the Presence of Tumor Changes

Purpose: To compare deformable image registration (DIR) accuracy and precision for normal and tumor tissues in head and neck cancer patients during the course of radiation therapy (RT). Methods and Materials: Thirteen patients with oropharyngeal tumors, who underwent submucosal implantation of small gold markers (average 6, range 4-10) around the tumor and were treated with RT were retrospectively selected. Two observers identified 15 anatomical features (landmarks) representative of normal tissues in the planning computed tomography (pCT) scan and in weekly cone beam CTs (CBCTs). Gold markers were digitally removed after semiautomatic identification in pCTs and CBCTs. Subsequently, landmarks and gold markers on pCT were propagated to CBCTs, using a b-spline-based DIR and, for comparison, rigid registration (RR). To account for observer variability, the pair-wise difference analysis of variance method was applied. DIR accuracy (systematic error) and precision (random error) for landmarks and gold markers were quantified. Time trend of the precisions for RR and DIR over the weekly CBCTs were evaluated. Results: DIR accuracies were submillimeter and similar for normal and tumor tissue. DIR precision (1 SD) on the other hand was significantly different (P<.01), with 2.2 mm vector length in normal tissue versus 3.3 mm in tumor tissue. No significant time trend in DIR precision was found for normal tissue, whereas in tumor, DIR precision was significantly (P<.009) degraded during the course of treatment by 0.21 mm/week. Conclusions: DIR for tumor registration proved to be less precise than that for normal tissues due to limited contrast and complex non-elastic tumor response. Caution should therefore be exercised when applying DIR for tumor changes in adaptive procedures.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2015

Lung Texture in Serial Thoracic Computed Tomography Scans: Correlation of Radiomics-based Features With Radiation Therapy Dose and Radiation Pneumonitis Development

Purpose: To assess the relationship between radiation dose and change in a set of mathematical intensity- and texture-based features and to determine the ability of texture analysis to identify patients who develop radiation pneumonitis (RP). Methods and Materials: A total of 106 patients who received radiation therapy (RT) for esophageal cancer were retrospectively identified under institutional review board approval. For each patient, diagnostic computed tomography (CT) scans were acquired before (0-168 days) and after (5-120 days) RT, and a treatment planning CT scan with an associated dose map was obtained. 32- × 32-pixel regions of interest (ROIs) were randomly identified in the lungs of each pre-RT scan. ROIs were subsequently mapped to the post-RT scan and the planning scan dose map by using deformable image registration. The changes in 20 feature values (ΔFV) between pre- and post-RT scan ROIs were calculated. Regression modeling and analysis of variance were used to test the relationships between ΔFV, mean ROI dose, and development of grade ≥2 RP. Area under the receiver operating characteristic curve (AUC) was calculated to determine each feature's ability to distinguish between patients with and those without RP. A classifier was constructed to determine whether 2- or 3-feature combinations could improve RP distinction. Results: For all 20 features, a significant ΔFV was observed with increasing radiation dose. Twelve features changed significantly for patients with RP. Individual texture features could discriminate between patients with and those without RP with moderate performance (AUCs from 0.49 to 0.78). Using multiple features in a classifier, AUC increased significantly (0.59-0.84). Conclusions: A relationship between dose and change in a set of image-based features was observed. For 12 features, ΔFV was significantly related to RP development. This study demonstrated the ability of radiomics to provide a quantitative, individualized measurement of patient lung tissue reaction to RT and assess RP development.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2017

Evaluation of a Machine-Learning Algorithm for Treatment Planning in Prostate Low-Dose-Rate Brachytherapy

Purpose: This work presents the application of a machine learning (ML) algorithm to automatically generate high-quality, prostate low-dose-rate (LDR) brachytherapy treatment plans. The ML algorithm can mimic characteristics of preoperative treatment plans deemed clinically acceptable by brachytherapists. The planning efficiency, dosimetry, and quality (as assessed by experts) of preoperative plans generated with an ML planning approach was retrospectively evaluated in this study. Methods and Materials: Preimplantation and postimplantation treatment plans were extracted from 100 high-quality LDR treatments and stored within a training database. The ML training algorithm matches similar features from a new LDR case to those within the training database to rapidly obtain an initial seed distribution; plans were then further fine-tuned using stochastic optimization. Preimplantation treatment plans generated by the ML algorithm were compared with brachytherapist (BT) treatment plans in terms of planning time (Wilcoxon rank sum, α = 0.05) and dosimetry (1-way analysis of variance, α = 0.05). Qualitative preimplantation plan quality was evaluated by expert LDR radiation oncologists using a Likert scale questionnaire. Results: The average planning time for the ML approach was 0.84 ± 0.57 minutes, compared with 17.88 ± 8.76 minutes for the expert planner (P=.020). Preimplantation plans were dosimetrically equivalent to the BT plans; the average prostate V150% was 4% lower for ML plans (P=.002), although the difference was not clinically significant. Respondents ranked the ML-generated plans as equivalent to expert BT treatment plans in terms of target coverage, normal tissue avoidance, implant confidence, and the need for plan modifications. Respondents had difficulty differentiating between plans generated by a human or those generated by the ML algorithm. Conclusions: Prostate LDR preimplantation treatment plans that have equivalent quality to plans created by brachytherapists can be rapidly generated using ML. The adoption of ML in the brachytherapy workflow is expected to improve LDR treatment plan uniformity while reducing planning time and resources.

62 RADIOLOGY AND NUCLEAR MEDICINE↗