Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Statistical techniques”

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 577 records · Page 32

An Assessment of Non-Powered Dam Hydropower Development Opportunities in the United States

Retrofitting non-powered dams (NPDs) to add hydropower offers multiple benefits over traditional, new hydropower development. Since much of the civil works infrastructure already exists, many retrofits require minimal new construction and can leverage existing water conveyances and discharge for generation. Although NPDs vary considerably in terms of their defining characteristics, identifying similarities helps support targeted investment and find opportunities to develop solutions for common challenges. With the publication of NPD-related data in 2022 (Hansen et al. 2022b), the US Department of Energy and relevant stakeholders are equipped with key information across roughly 89,000 US NPDs. These data represent the best-available information to date and extend beyond previous assessments of potential power capacity by describing design, operational, socioeconomic, environmental aspects of NPDs (Hadjerioua et al. 2012). To capitalize on these efforts to improve breadth and depth of NPD data access, this study uses a data-driven approach to assess NPD hydropower development opportunities in the United States. To help describe project feasibility drivers for NPDs, recent NPD retrofits were reviewed to examine variability with respect to a variety of characteristics. Based on the assessment of recent retrofits and data availability, several attributes were selected to characterize the remaining NPD population: (1) owner type, (2) 30% exceedance flow (a common design flow that describes the flow that is exceeded by 30% of the flow in the record), (3) hydraulic head, and (4) maximum reservoir storage. These four characteristics were used as inputs to a statistical clustering analysis (a common technique for grouping individuals of a population based on similarities or how closely associated individuals are to one another) of a set of 2,709 NPDs with at least 100 kW potential capacity and recently retrofit dams with available data. With the large dataset of NPDs, the clusters help describe how the population breaks down into different types (i.e., how many types of dams there are and what portion of the population belongs to each cluster).

13 HYDRO ENERGY↗

LENS: Learning Enabled Network Synthesis

RTRC and UMD have developed novel machine learning based methods under the ARPA-E DIFFERENTIATE program for rapid acceleration of hypothesis generation in complex architecture design spaces involving both discrete choices of component inclusion and interconnection and continuous parametric decisions. The project named Learning Enabled Network Synthesis (LENS) further demonstrated the developed methods on challenging electrical power converter design problems by identifying the most suitable circuit topologies and simultaneously selecting the most appropriate components to achieve optimized design of power converter with improved performances. We demonstrated that LENS could enable exploration of very large design space of circuit topologies and components by addressing the limitations of conventional design process in non-linear, high switching speed, multi-dimensional power converter design and optimization. The key innovation developed in LENS is the seamless integration of statistical learning and logical reasoning techniques and building on the individual strengths of these techniques for rapid hypothesis discovery. The main component of LENS comprises of: 1) Graph Reasoning Engine (GRE) to enforce composition rules that rapidly reject all discrete architectures that are composed incorrectly and generates an adaptive database of feasible designs which can be used by ML modules, 2) Graph Generative Learning module which is a deep neural network based generative model for graph architectures which can enable design space exploration beyond the dataset generated by the GRE, 3) Graph Reduced Order Model (ROM) for graph domains for accelerating computation of output metrics, and 4) Active learning and Rule Discovery module for sample efficient learning and extracting logical rules from the learned ML models which will be integrated in the GRE to enhance the filtering effectiveness. LENS approach can be applied to any design domains where designs can be represented as multi-attribute graphs. The LENS team integrated the various technical innovations listed above into an optimization pipeline and exercised the optimization pipeline on the converter design problem. The LENS project demonstrated that the developed AI/ML technologies can be used to generate novel converter circuits >45x faster than experts on chosen use-cases. This can enable faster design space exploration and identification of new designs which are not considered by experts due to the increasing design space complexity. This has significant potential impact on the public and energy needs of the country. It is currently estimated that 30% of all electrical powers generated passes through power converters. The future estimate is that 80% of all power generated would be passing through converters. LENS fills a critical gap in this space since by accelerating the design process the designers would be able to generate more efficient converters which can lead to significant energy savings for the country.

42 ENGINEERING↗

Graph Neural Networks for Parameterized Quantum Circuits Expressibility Estimation (Rev.1)

Parameterized quantum circuits (PQCs) are fundamental to quantum machine learning (QML), quantum optimization, and variational quantum algorithms (VQAs). The expressibility of PQCs is a measure that determines their capability to harness the full potential of the quantum state space. It is thus a crucial guidepost to know when selecting a particular PQC ansatz. However, the existing technique for expressibility computation through statistical estimation requires a large number of samples, which poses significant challenges due to time and computational resource constraints. This paper introduces a novel approach for expressibility estimation of PQCs using Graph Neural Networks (GNNs). We demonstrate the predictive power of our GNN model with a dataset consisting of 25,000 samples from the noiseless IBM QASM Simulator and 12,000 samples from three distinct noisy quantum backends. The model accurately estimates expressibility, with root mean square errors (RMSE) of 0.05 and 0.06 for the noiseless and noisy backends, respectively. We compare our model’s predictions with reference circuits from Sim et al. and IBM Qiskit’s hardwareefficient ansatz sets to further evaluate our model’s performance. Our experimental evaluation in noiseless and noisy scenarios reveals a close alignment with ground truth expressibility values, highlighting the model’s efficacy. Moreover, our model exhibits promising extrapolation capabilities, predicting expressibility values with low RMSE for out-of-range qubit circuits trained solely on only up to 5-qubit circuit sets. This work thus provides a reliable means of efficiently evaluating the expressibility of diverse PQCs on noiseless simulators and hardware.

97 MATHEMATICS AND COMPUTING↗

Study of the heavy flavour fractions in z+jets events from $p\bar{p}$ collisions at energy √s = 1.96 TeV with the CDF II detector at the Tevatron collider

The Standard Model of field and particles is the theory that provides the best description of the known phenomenology of the particle physics up to now. Data collected in the last years, mainly by the experiments at the big particle accelerators (SPS, LEP, TEVATRON, HERA, SLAC), allowed to test the agreement between measurements and theoretical calculations with a precision of 10 -3 / 10 -4 . The Standard Model is a Quantum Field Theory based on the gauge symmetry group SU(3) C x SU(2) L x U(1) Y , with spontaneous symmetry breaking. This gauge group includes the color symmetry group of the strong interaction, SU(3) C , and the symmetry group of the electroweak interactions, SU(2) L x U(1) Y . The formulation of the Standard Model as a gauge theory guarantees its renormalizability, but forbids explicit mass terms for fermions and gauge bosons. The masses of the particles are generated in a gauge-invariant way by the Higgs Mechanism via a spontaneous breaking of the electroweak symmetry. This mechanism also implies the presence of a massive scalar particle in the mass spectrum of the theory, the Higgs boson. This particle is the only one, among the basic elements for the minimal formulation of the Standard Model, to have not been confirmed by the experiments yet. For this reason in the last years the scientific community has been focusing an increasing fraction of its efforts on the search of the Higgs boson. The mass of the Higgs boson is a free parameter of the Standard Model, but the unitarity of the theory requires values not higher than 1 TeV and the LEP experiments excluded values smaller than 115 GeV. To explore this range of masses is under construction at CERN the Large Hadron Collider (LHC), a proton-proton collider with a center of mass energy of 14 TeV and a 10 34 cm -2 s -1 peak luminosity. According to the present schedule, this machine will start to provide collisions for the experiments at the end of 2008. In the meanwhile the only running accelerator able to provide collisions suitable for the search of the Higgs boson is the Tevatron at Fermilab, a proton-antiproton collider with a center of mass energy of 1.96 TeV working at 3 • 10 32 cm -2 s -1 peak luminosity. These features make the Tevatron able for the direct search of the Higgs boson in the 115-200 GeV mass range. Since the coupling of the Higgs boson is proportional to the masses of the particles involved, the decay in b{bar b} has the largest branching ratio for Higgs mass < 135 GeV and thus the events Z/W + $b\bar{b}$ are the main background to the Higgs signal in the most range favored by Standard Model fits. In this thesis a new technique to identify Heavy Flavour quarks inside high - P T jets is applied to events with a reconstructed Z boson to provide a measurement of the Z+b and Z+c inclusive cross sections. The study of these channels represent also a test of QCD in high transferred momentum regime, and can provide information on proton pdf. This new Heavy Flavour identication technique (tagger) provides an increased statistical separation between b, c and light flavours, using a new vertexing algorithm and a chain of artificial Neural Networks to exploit as much information as possible in each event. For this work I collaborated with the Universita di Roma 'La Sapienza' group working in the CDF II experiment at Tevatron, that has at first developed this tagger. After a brief theoretical introduction (chapter 1) and a description of the experimental apparatus (chapter 2), the tagger itself and its calibration procedure are described in chapter 3 and 4. The chapter 5 is dedicated to the event selection and the chapter 6 contains the results of the measurement and the study of the systematic errors.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Angular Correlation Function Estimators Accounting for Contamination from Probabilistic Distance Measurements

With the advent of surveys containing millions to billions of galaxies, it is imperative to develop analysis techniques that utilize the available statistical power. In galaxy clustering, even small sample contamination arising from distance uncertainties can lead to large artifacts, which the standard estimator for two-point correlation functions does not account for. We first introduce a formalism, termed decontamination, that corrects for sample contamination by utilizing the observed cross-correlations in the contaminated samples; this corrects any correlation function estimator for contamination. Using this formalism, we present a new estimator that uses the standard estimator to measure correlation functions in the contaminated samples but then corrects for contamination. We also introduce a weighted estimator that assigns each galaxy a weight in each redshift bin based on its probability of being in that bin. We demonstrate that these estimators effectively recover the true correlation functions and their covariance matrices. Our estimators can correct for sample contamination caused by misclassification between object types as well as photometric redshifts; they should be particularly helpful for studies of galaxy evolution and baryonic acoustic oscillations, where forward modeling the clustering signal using the contaminated redshift distribution is undesirable.

79 ASTRONOMY AND ASTROPHYSICS↗

Applications of High Resolution Laser Induced Breakdown Spectroscopy for Environmental and Biological Samples

This chapter details the application of LIBS in a number of environmental areas of research such as carbon sequestration and climate change. LIBS has also been shown to be useful in other high resolution environmental applications for example, elemental mapping and detection of metals in plant materials. LIBS has also been used in phytoremediation applications. Other biological research involves a detailed understanding of wood chemistry response to precipitation variations and also to forest fires. A cross-section of Mountain pine (pinceae Pinus pungen Lamb.) was scanned using a translational stage to determine the differences in the chemical features both before and after a fire event. Consequently, by monitoring the elemental composition pattern of a tree and by looking for abrupt changes, one can reconstruct the disturbance history of a tree and a forest. Lastly we have shown that multivariate analysis of the LIBS data is necessary to standardize the analysis and correlate to other standard laboratory techniques. LIBS along with multivariate statistical analysis makes it a very powerful technology that can be transferred from laboratory to field applications with ease.

Martin, Madhavi Z↗

Anatomical Fraction segmentation in the biomass bales

The U.S. Department of Energy Bioenergy Technologies Office (BETO) is committed to the development of sustainable, nationwide, commercial biofuel production to displace petroleum-derived fuels, increase domestic energy production, and encourage the creation of a domestic bioenergy and bioproducts industry. Operating a commercial-scale bioenergy operation requires significant technological advancements for determining biomass feedstock quality during the preprocessing stage. Usage of non-food feedstocks- e.g., corn-stover, pine residue, and forest residue - at the biorefineries reduces strain on the food supply chain. But the non-food feedstock heterogeneity- physical size, shape, and chemical composition-poses a significant challenge during milling, conveyance, feeding, and biofuel conversion processes. 3D X-Ray imaging (CT) is capable of distinguishing material features by detecting density and compositional differences. But the inherent complexity introduced during harvesting and bailing makes the reconstruction and interpretation of baled biomass materials from x-ray data time consuming, laborious, and expensive. Feedstocks are low-density materials that produce small contrast differences in the x-ray images. This paper focuses on using the shape and texture properties, using 3D image processing techniques like 3D skeletonization, directional statistics to characterize and extract volumetric content of the different tissue samples in the biomass bales.

09 BIOMASS FUELS↗

Understanding Manufacturing Scale-Up of Plasma Sprayed Zr Barrier Coatings on U-10Mo Fuels

The United States government is committed to further strengthening nuclear security and nonproliferation in order to reduce the threat of terrorists acquiring nuclear material. To meet this important mission, the US Department of Energy/National Nuclear Security Agency's Office of Material Management and Minimization (M{sup 3}) Reactor Conversion Program minimizes the use of highly enriched uranium (HEU) in civilian applications by working with governments and facilities around the world to convert research reactors to the use of non-weapons-usable low-enriched uranium (LEU) fuels. Each reactor converted not only eliminates the need for HEU at civilian sites, but also reduces the amount of HEU being manufactured, stored, and transported, where it is at its most vulnerable. In instances where suitable LEU fuels do not exist for particular reactors to convert, the M{sup 3} Reactor Conversion Program (CONVERT) contributes to the development of new LEU fuels. Currently, the M{sup 3} Reactor Conversion Program is committed to developing a high-density U-Mo LEU fuel that will be able to convert high performance research reactors while establishing an efficient and economic fabrication capability to make the fuel. To that end, Los Alamos National Laboratory (LANL) has been tasked with developing a method to affix a Zirconium fuel diffusion barrier via plasma spraying that will prevent a reaction between the U-Mo fuel and the aluminum cladding. This report discusses the transitions from development of the plasma spray deposition process to a scientific understanding of pseudo-production conditions and quantification of coating layer properties, characterization, reproducibility, etc. Understanding quality assurance and its applications to experiment work is non-trivial; statistical analysis of various measurement techniques not only improve quality, but will allow for scale up of manufacturing.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Capturing dynamics on multiple timescales: a hybrid approach for cluttered electromagnetic data

Many problems in electromagnetic signal analysis exhibit dynamics on a wide range of time scales against nonstationary clutter and noise. We consider a problem in which the relevant time scales can range from nanoseconds to hours or days (12 or 13 orders of magnitude). We present a hybrid algorithm currently designed to capture the dynamic behavior at scales from nanoseconds to milliseconds (6 orders of magnitude) while remaining robust to clutter and noise. We draw from techniques of adaptive feature extraction, statistical machine learning, and discrete process modeling and present results on a simulated multimode problem. Our goals are to find a representation of the signal that allows us to identify which pulses were produced by a target emitter and to determine the operational mode of the target.

97 MATHEMATICS AND COMPUTING↗

A statistical method for retrospective cardiac and respiratory motion gating of interventional cardiac x-ray images

Purpose: Image-guided cardiac interventions involve the use of fluoroscopic images to guide the insertion and movement of interventional devices. Cardiorespiratory gating can be useful for 3D reconstruction from multiple x-ray views and for reducing misalignments between 3D anatomical models overlaid onto fluoroscopy. Methods: The authors propose a novel and potentially clinically useful retrospective cardiorespiratory gating technique. The principal component analysis (PCA) statistical method is used in combination with other image processing operations to make our proposed masked-PCA technique suitable for cardiorespiratory gating. Unlike many previously proposed techniques, our technique is robust to varying image-content, thus it does not require specific catheters or any other optically opaque structures to be visible. Therefore, it works without any knowledge of catheter geometry. The authors demonstrate the application of our technique for the purposes of retrospective cardiorespiratory gating of normal and very low dose x-ray fluoroscopy images. Results: For normal dose x-ray images, the algorithm was validated using 28 clinical electrophysiology x-ray fluoroscopy sequences (2168 frames), from patients who underwent radiofrequency ablation (RFA) procedures for the treatment of atrial fibrillation and cardiac resynchronization therapy procedures for heart failure. The authors established end-systole, end-expiration, and end-inspiration success rates of 97.0%, 97.9%, and 97.0%, respectively. For very low dose applications, the technique was tested on ten x-ray sequences from the RFA procedures with added noise at signal to noise ratio (SNR) values of√(5)0, √(1)0, √(8), √(6), √(5), √(2), and √(1) to simulate the image quality of increasingly lower dose x-ray images. Even at the low SNR value of √(2), representing a dose reduction of more than 25 times, gating success rates of 89.1%, 88.8%, and 86.8% were established. Conclusions: The proposed technique can therefore extract useful information from interventional x-ray images while minimizing exposure to ionizing radiation.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Optimization Studies of Radiation Shielding for the PIP-II Project at Fermilab

The PIP-II project at Fermilab, which includes an 800-MeV superconducting LINAC, demands rigorous radiation shielding optimization to meet safety requirements. We updated the MARS geometry model to reflect new magnet and collimator designs and introduced high-resolution detector planes to better capture radiation field distributions. To overcome the significant computational demands, we implemented a well-known branching technique that drastically reduced simulation runtimes while maintaining statistical integrity. This was achieved through particle splitting and the application of Russian Roulette techniques. Additionally, new graphical tools were created to streamline data visualization and MARS code usability.

Makovec, Alajos [Fermilab] (ORCID:0000000286157492↗

Dosimetric Comparison of RapidArc with Fixed Gantry Intensity-Modulated Radiotherapy Treatment for Multiple Liver Metastases Radiotherapy

We wanted to compare the dosimetric difference and treatment efficiency of RapidArc and fixed gantry intensity-modulated radiotherapy treatment (IMRT) for multiple liver metastases. Computed tomography datasets of 10 patients were studied retrospectively. IMRT plans were generated using 5 fields and RapidArc using either 1 or 2 arcs. The dose distribution of planning target volume (PTV), organs at risk (OARs), and the normal tissue were compared. Monitor units and treatment time were scored to measure expected treatment efficiency. Both RapidArc and IMRT plans resulted in equivalent target coverage. There was no statistically significant difference for the maximum and the minimum dose of PTV. RapidArc plans achieved an improved conformity index compared with IMRT (RA1 = 1.68 {+-} 0.27, RA2 = 1.61 {+-} 0.25, IMRT = 1.80 {+-} 0.37). For OARs, all techniques respected planning objectives. RapidArc plans had a lower dose in V{sub 40} of small bowel than IMRT, but were higher in mean dose of kidneys. Concerning the V{sub 5}, V{sub 10}, and V{sub 15} of healthy tissue, RapidArc plans were higher than IMRT. However, the V{sub 20}, V{sub 25}, and V{sub 30} of healthy tissue in RapidArc plans were lower than IMRT. Monitor units per fraction of RapidArc plans were about 40% or 46% of IMRT. Compared with IMRT plans, treatment time of RapidArc plans were reduced by 60% or 70%. All techniques respected planning objectives. RapidArc showed statistical improvements in conformity index and healthy tissue sparing with uncompromised target coverage. This, in combination with fewer monitor units and short delivery time, can lead to clinically significant advances for the treatment of multiple liver metastases.

61 RADIATION PROTECTION AND DOSIMETRY↗

Correction of Dual-PRF Doppler Velocity Outliers in the Presence of Aliasing

Abstract In Doppler weather radars, the presence of unfolding errors or outliers is a well-known quality issue for radial velocity fields estimated using the dual–pulse repetition frequency (PRF) technique. Postprocessing methods have been developed to correct dual-PRF outliers, but these need prior application of a dealiasing algorithm for an adequate correction. This paper presents an alternative procedure based on circular statistics that corrects dual-PRF errors in the presence of extended Nyquist aliasing. The correction potential of the proposed method is quantitatively tested by means of velocity field simulations and is exemplified in the application to real cases, including severe storm events. The comparison with two other existing correction methods indicates an improved performance in the correction of clustered outliers. The technique proposed is well suited for real-time applications requiring high-quality Doppler radar velocity fields, such as wind shear and mesocyclone detection algorithms, or assimilation in numerical weather prediction models.

58 GEOSCIENCES↗

A Parameter-masked Mock Data Challenge for Beyond-two-point Galaxy Clustering Statistics

The past few years have seen the emergence of a wide array of novel techniques for analyzing high-precision data from upcoming galaxy surveys, which aim to extend the statistical analysis of galaxy clustering data beyond the linear regime and the canonical two-point (2pt) statistics. We test and benchmark some of these new techniques in a community data challenge named “Beyond-2pt,” initiated during the Aspen 2022 Summer Program “Large-Scale Structure Cosmology beyond 2-Point Statistics,” whose first round of results we present here. The challenge data set consists of high-precision mock galaxy catalogs for clustering in real space, in redshift space, and on a light cone. Participants in the challenge have developed end-to-end pipelines to analyze mock catalogs and extract unknown (“masked”) cosmological parameters of the underlying ΛCDM models with their methods. The methods represented are density-split clustering, nearest neighbor statistics, BACCO power spectrum emulator, void statistics, LEFTfield field-level inference using effective field theory (EFT), and joint power spectrum and bispectrum analyses using both EFT and simulation-based inference. In this work, we review the results of the challenge, focusing on problems solved, lessons learned, and future research needed to perfect the emerging beyond-2pt approaches. The unbiased parameter recovery demonstrated in this challenge by multiple statistics and the associated modeling and inference frameworks supports the credibility of cosmology constraints from these methods. The challenge data set is publicly available, and we welcome future submissions from methods that are not yet represented.

Krause, Elisabeth [Univ. of Arizona, Tucson, AZ (U↗

An optimized posterior axillary boost technique in radiation therapy to supraclavicular and axillary lymph nodes: A comparative study

To assess the advantages of an optimized posterior axillary (AX) boost technique for the irradiation of supraclavicular (SC) and AX lymph nodes. Five techniques for the treatment of SC and levels I, II, and III AX lymph nodes were evaluated for 10 patients selected at random: a direct anterior field (AP); an anterior to posterior parallel pair (AP-PA); an anterior field with a posterior axillary boost (PAB); an anterior field with an anterior axillary boost (AAB); and an optimized PAB technique (OptPAB). The target coverage, hot spots, irradiated volume, and dose to organs at risk were evaluated and a statistical analysis comparison was performed. The AP technique delivered insufficient dose to the deeper AX nodes. The AP-PA technique produced larger irradiated volumes and higher mean lung doses than the other techniques. The PAB and AAB techniques originated excessive hot spots in most of the cases. The OptPAB technique produced moderate hot spots while maintaining a similar planning target volume (PTV) coverage, irradiated volume, and dose to organs at risk. This optimized technique combines the advantages of the PAB and AP-PA techniques, with moderate hot spots, sufficient target coverage, and adequate sparing of normal tissues. The presented technique is simple, fast, and easy to implement in routine clinical practice and is superior to the techniques historically used for the treatment of SC and AX lymph nodes.

61 RADIATION PROTECTION AND DOSIMETRY↗

Decoding defect statistics from diffractograms via machine learning

Abstract Diffraction techniques can powerfully and nondestructively probe materials while maintaining high resolution in both space and time. Unfortunately, these characterizations have been limited and sometimes even erroneous due to the difficulty of decoding the desired material information from features of the diffractograms. Currently, these features are identified non-comprehensively via human intuition, so the resulting models can only predict a subset of the available structural information. In the present work we show (i) how to compute machine-identified features that fully summarize a diffractogram and (ii) how to employ machine learning to reliably connect these features to an expanded set of structural statistics. To exemplify this framework, we assessed virtual electron diffractograms generated from atomistic simulations of irradiated copper. When based on machine-identified features rather than human-identified features, our machine-learning model not only predicted one-point statistics (i.e. density) but also a two-point statistic (i.e. spatial distribution) of the defect population. Hence, this work demonstrates that machine-learning models that input machine-identified features significantly advance the state of the art for accurately and robustly decoding diffractograms.

36 MATERIALS SCIENCE↗

Statistics of Fractionalized Excitations through Threshold Spectroscopy

We show that neutral anyonic excitations have a signature in spectroscopic measurements of materials: The low-energy onset of spectral functions near the threshold follows universal power laws with an exponent that depends only on the statistics of the anyons. This provides a route, using experimental techniques such as neutron scattering and tunneling spectroscopy, for detecting anyonic statistics in topologically ordered states such as gapped quantum spin liquids and hypothesized fractional Chern insulators. Our calculations also explain some recent theoretical results in spin systems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Evaluation of a Single-Isocenter Technique for Axillary Radiotherapy in Breast Cancer

The aim of this study was to develop a technique for axillary radiotherapy that minimizes the risk of radiation-induced damage to the surrounding normal tissue (i.e., arm, shoulder, lung, esophagus, and spinal cord) while keeping the risk of a nodal recurrence to a minimum. A planning study was performed in 20 breast cancer patients. The target volume of the axillary treatment encompassed the periclavicular and axillary lymph node areas. The 3-dimensional (3D) computed tomography (CT) information in this study was used to outline the lymph node areas and the organs at risk (i.e., the esophagus, spinal cord, brachial plexus, and lung). A conventional AP-PA technique (with a transmission plate placed in the AP beam) was evaluated. In addition, a new single-isocenter technique consisting of AP/PA fields using a gantry rotation of {+-}20 deg. and a medial AP segment was developed. Both techniques were compared by evaluation of the calculated dose distributions and the dose-volume histograms of the target volume and surrounding organs at risk. The field borders and humeral shielding were redefined based on the 3D anatomical references. Adapting the humeral shielding reduced the irradiated volume by 19% and might contribute to a reduction of the incidence of arm edema and impairment of shoulder function. The maximum radiation dose in the esophagus and spinal cord was reduced by more than 50% using the single-isocenter technique. The difference between both techniques with respect to the mean doses in the target volume and lung, and the maximum dose in brachial plexus, was not statistically significant. Moreover, the single-isocenter technique allowed a fast and easy treatment preparation and reduced the execution time considerably (with approximately 10 minutes per fraction)

62 RADIOLOGY AND NUCLEAR MEDICINE↗