Search NASA⌕ Search

SEARCH · Search NASA

Results for “control methods”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 559 records · Page 31

Engineering Considerations for Remote Refabrication of EBR-II Fuel Elements

The development of remote controlled methods for manufacturing EBR-II fuel elements was influenced by many interacting factors. Radiation levels within the process cell have been predicted to range from 10 3 to 10 7 rad per hour. Radiation damage to organic lubricant, electrical insulations, elastic seals, and protective coatings precludes the use of many standard machine components. Heat generated in the fuel by absorbed radiation makes forced cooling necessary in many operations. Oxygen must be exluded from all operations where the fuel is exposed. Equipment must be designed for remote maintenance and component replacement within the limitation of available manipulators. The EBR-II fuel consisted of fissium alloy pins sodium bonded in stainless steel tubes. Precision casting was chosen as the basis for refabricating the fuel pins. Remote controlled equipment was developed to cast, assemble, and inspect the EBR-II fuel elements. Radiation resistant, plug-in machine components were developed to give reasonable life expectancy and to allow remote maintenance and replacement.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Virtual to Physical: Reinforcement Learning to Optimize SNS Particle Accelerator Controls

Complex accelerators must have control systems that can handle dynamic nonlinear environments. This makes traditional control methods unsuitable as they can struggle to adapt to these uncertainties. This provides an ideal environment for reinforcement learning algorithms as they are adaptable and generalizable. We present a reinforcement learning pipeline that can effectively handle the dynamics of a complex accelerator. We test and prove our pipelines capabilities on multiple environments including the Spallation Neutron Source (SNS) and the Beam Test Facility (BTF) at Oakridge National Lab (ORNL). Due to the limited time available to train an online algorithm like reinforcement learning on a real accelerator, we utilize a virtual twin accelerator (VIRAC) developed by ORNL to pretrain the policy and show its ability to converge in the virtual environment. We then test the adaptability of the pretrained RL model by applying it on the real accelerator and comparing the results. Utilizing our Scientific Optimization and Controls Toolkit (SOCT) and open-source standards such as Gymnasium we create and solve for a MEBT orbit correction problem in the SNS and an emittance maximization problem in the BTF. We show how Twin Delayed Deep Deterministic Policy Gradient (TD3) can solve this optimization environment in the virtual accelerator and transfer this policy onto the real accelerator for inference and model retraining. We show how reinforcement learning can be utilized as a control system for complex accelerators and provide a model pipeline for how an implementation performs and can be adapted to new accelerator control problems.

Kasparian, Armen [Thomas Jefferson National Accele↗

FINAL SCIENTIFIC/TECHNICAL REPORT

Dynalene Inc has developed and patented a fuel cell coolant with the help of DOE SBIR Phase I and Phase II funding (Project DE-FG02-04ER83884). However, this coolant could only be produced in lab scale (500 ml to 2 L) due to problems in the optimization and scale-up of a nanoparticle ingredient. This project optimized the nanoparticle production process in 10 L and 100 L reactors (which translates to about 5000 gallons of coolant), optimized the filtration process for the nanoparticles, and develop a high throughput production as well as quality control method for the final coolant formulation. Scale-up of nanoparticle synthesis (using emulsion polymerization) is an extremely challenging task. Dynalene researchers, in collaboration with a university partner, identified all the parameters affecting the size, charge density and coagulation characteristics of the nanoparticles and then optimized these parameters to achieve the goals and the objectives of this project. Nanoparticle synthesis was demonstrated to be reproducible in the 10 L and 100 L scales.

08 HYDROGEN↗

Thermal storage system with coupled tanks

Methods and apparatus are disclosed for high-efficiency thermal storage with a fluid-filled “battery” tank positioned within a fluid-filled “reservoir” tank. Fluid loops couple the tanks to a heat pump and a building. The heat pump can charge the battery tank or deliver thermal energy (cold or heat) to a building, using the reservoir tank or ambient air as a thermal energy source. The battery tank can discharge energy to the building jointly with the heat pump or, at periods of peak electricity usage, with the heat pump switched off. Operating modes allow significant savings in electricity usage and mitigate the “duck curve.” Low duty cycle usage of the reservoir enables efficient underground thermal storage with less digging than conventional geothermal technologies. Additional efficiency is achieved with phase change materials installed inside a tank or in a tank wall, providing temperature regulation. Control methods are disclosed.

15 GEOTHERMAL ENERGY↗

Techno-Economic Analysis for the Addition of a Thermal Energy Storage System to a Central Plant

Increasing energy demand and rising peak loads present significant challenges for energy management in commercial and institutional settings. As climate change drives greater cooling needs, central plants must navigate the complex tradeoffs between operational efficiency, cost control, and grid stability. Thermal energy storage (TES) systems offer a viable solution by shifting energy consumption from peak to off-peak periods, thereby reducing peak demand, lowering utility expenses, and improving grid resilience. However, the success of TES implementation hinges on appropriate system sizing, effective control strategies, and alignment with local utility rate structures. This article presents a techno-economic analysis of integrating a chilled water TES system into the central plant at California State University, Dominguez Hills. Drawing on historical load profiles and utility tariffs, we assess three TES sizing approaches and their corresponding control strategies from both energy and economic perspectives. This article utilizes a model-based approach to assess the impact of TES sizing and control strategies on the techno-economic feasibility of integrating TES into an existing central plant. The models employed for this analysis were calibrated using 4 years of historical data. Here, the results demonstrated that utility tariffs and the campus's operational profiles dictate the most feasible sizing and control methods. The findings offer valuable insights for institutions and commercial building managers exploring sustainable energy solutions. By demonstrating how optimized TES strategies can improve operational efficiency while achieving financial savings, this study highlights the potential for TES to align performance with cost effectiveness in real-world applications.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Preoperative Single Fraction Partial Breast Radiotherapy for Early-Stage Breast Cancer

Purpose: Several recent series evaluating external beam accelerated partial breast irradiation (PBI) have reported adverse cosmetic outcomes, possibly related to large volumes of normal tissue receiving near-prescription doses. We hypothesized that delivery of external beam PBI in a single fraction to the preoperative tumor volume would be feasible and result in a decreased dose to the uninvolved breast compared with institutional postoperative PBI historical controls. Methods and Materials: A total of 17 patients with unifocal Stage T1 breast cancer were identified. Contrast-enhanced subtraction magnetic resonance images were loaded into an Eclipse treatment planning system and used to define the target volumes. A 'virtual plan' was created using four photon beams in a noncoplanar beam arrangement and optimized to deliver 15 Gy to the planning target volume. Results: The median breast volume was 1,713 cm{sup 3} (range: 1,014-2,140), and the median clinical target volume was 44 cm{sup 3} (range: 26-73). In all cases, 100% of the prescription dose covered 95% of the clinical target volume. The median conformity index was 0.86 (range: 0.70-1.12). The median percentage of the ipsilateral breast volume receiving 100% and 50% of the prescribed dose was 3.8% (range: 2.2-6.9) and 13.3% (range: 7.5-20.8) compared with 18% (range: 3-42) and 53% (range: 24-65) in the institutional historical controls treated with postoperative external beam PBI (p = .002). The median maximum skin dose was 9 Gy. The median dose to 1 and 10 cm{sup 3} of skin was 6.7 and 4.9 Gy. The doses to the heart and ipsilateral lung were negligible. Conclusion: Preoperative PBI resulted in a substantial reduction in ipsilateral breast tissue dose compared with postoperative PBI. The skin dose appeared reasonable, given the small volumes. A prospective Phase I trial evaluating this technique is ongoing.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Carbon Ion Radiotherapy for Unresectable Retroperitoneal Sarcomas

Purpose: To evaluate the applicability of carbon ion radiotherapy (CIRT) for unresectable retroperitoneal sarcomas with regard to normal tissue morbidity and local tumor control. Methods and Materials: From May 1997 to February 2006, 24 patients (17 male and 7 female) with unresectable retroperitoneal sarcoma received CIRT. Age ranged from 16 to 77 years (median, 48.6 years). Of the patients, 16 had primary disease and 8 recurrent disease. Histologic diagnoses were as follows: malignant fibrous histiocytoma in 6, liposarcoma in 3, malignant peripheral nerve sheath tumor in 3, Ewing/primitive neuroectodermal tumor (PNET) in 2, and miscellaneous in 10 patients. The histologic grades were as follows: Grade 3 in 15, Grade 2-3 in 2, Grade 2 in 3, and unknown in 4. Clinical target volumes ranged between 57 cm{sup 3} and 1,194 cm{sup 3} (median 525 cm{sup 3}). The delivered carbon ion dose ranged from 52.8 to 73.6 GyE in 16 fixed fractions over 4 weeks. Results: The median follow-up was 36 months (range, 6-143 months). The overall survival rates at 2 and 5 years were 75% and 50%, respectively. The local control rates at 2 and 5 years were 77% and 69%. No complications of the gastrointestinal tract were encountered. No other toxicity greater than Grade 2 was observed. Conclusions: Use of CIRT is suggested to be effective and safe for retroperitoneal sarcomas. The results obtained with CIRT were a good overall survival rate and local control, notwithstanding the fact that most patients were not eligible for surgical resection and had high-grade sarcomas.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Energy Efficient Removal of Volatile Organic Compounds (VOCs) and Organic Hazardous Air Pollutants (o-HAPs) from Industrial Waste Streams by Direct Electron Oxidation

This research program investigated and quantified the capability of direct electron beam destruction of volatile organic compounds and organic hazardous air pollutants in model industrial waste streams and calculated the energy savings that would be realized by the widespread adoption of the technology over traditional pollution control methods. Specifically, this research determined the quantity of electron beam dose required to remove 19 of the most important non-halogenated air pollutants from waste streams and constructed a technical and economic model for the implementation of the technology in key industries including petroleum refining, organic & solvent chemical production, food & beverage production, and forest & paper products manufacturing. Energy savings of 75 - 90% and green house gas reductions of 66 - 95% were calculated for the target market segments.

02 PETROLEUM↗

An Information Theoretic Approach to Identify Dominant Voltage Influencers for Unbalanced Distribution Systems

Smart distribution grid with multiple renewable energy sources can experience random voltage fluctuations due to variable generation, which may result in voltage violations. Traditional voltage control algorithms are inadequate to handle fast voltage variations. Therefore, new dynamic control methods are being developed that can significantly benefit from the knowledge of dominant voltage influencer (DVI) nodes. DVI nodes for a particular node of interest refer to nodes that have a relatively high impact on the voltage fluctuations at that node. Conventional power flow-based algorithms to identify DVI nodes are computationally complex, which limits their use in real-time applications. This paper proposes a novel information theoretic voltage influencing score (VIS) that quantifies the voltage influencing capacity of nodes with DERs/active loads in a three phase unbalanced distribution system. VIS is then employed to rank the nodes and identify the DVI set. VIS is derived analytically in a computationally efficient manner and its efficacy to identify DVI nodes is validated using the IEEE 37-node test system. It is shown through experiments that KL divergence and Bhattacharyya distance are effective indicators of DVI nodes with an identifying accuracy of more than 90%. Additionally, the computation burden is also reduced by an order of 5, thus providing the foundation for efficient voltage control.

42 ENGINEERING↗

Direct pulse-level compilation of arbitrary quantum logic gates on superconducting qutrits

Advanced simulations and calculations on quantum computers require high-fidelity implementations of quantum operations. The universal gateset approach builds complex unitaries from a small set of primitive gates, often resulting in a long gate sequence, which is typically a leading factor in the total accumulated error. Compiling a complex unitary for processors with higher-dimensional logical elements, such as qutrits, exacerbates the accumulated error per unitary, since an even longer gate sequence is required. Optimal control methods promise time- and resource-efficient compact gate sequences and, therefore, higher fidelity. These methods generate pulses that can directly implement any complex unitary on a quantum device. In this work, we demonstrate that any arbitrary qubit and qutrit gate can be realized with high fidelity, which can significantly reduce the length of a gate sequence. We generate and test pulses for a large set of randomly selected arbitrary unitaries on several quantum processing units (QPUs): the Lawrence Livermore National Laboratory Quantum Device and Integration Testbed’s (QuDIT’s) standard QPU and three of Rigetti’s QPUs: Ankaa-2, Ankaa-9Q-1, and Aspen-M-3. On the QuDIT platform’s standard QPU, the average fidelity of random qutrit gates is 97.9 ± 0.5% measured with conventional QPT and 98.8 ± 0.6% from QPT with gate folding. Rigetti’s Ankaa-2 achieves random qubit gates with an average fidelity of 98.4 ± 0.5% (conventional QPT) and 99.7 ± 0.1% (QPT with gate folding). On Ankaa-9Q-1 and Aspen-M-3, the average fidelities with conventional qubit QPT measurements were higher than 99% (see Appendix). Here we show that optimal control gates are robust to drift for at least 3 h and that the same calibration parameters can be used for all implemented gates. Our work promises that the calibration overheads for optimal control gates can be made small enough to enable efficient quantum circuits based on this technique.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Cervical Gross Tumor Volume Dose Predicts Local Control Using Magnetic Resonance Imaging/Diffusion-Weighted Imaging—Guided High-Dose-Rate and Positron Emission Tomography/Computed Tomography—Guided Intensity Modulated Radiation Therapy

Purpose: Magnetic resonance imaging/diffusion weighted-imaging (MRI/DWI)-guided high-dose-rate (HDR) brachytherapy and {sup 18}F-fluorodeoxyglucose (FDG) — positron emission tomography/computed tomography (PET/CT)-guided intensity modulated radiation therapy (IMRT) for the definitive treatment of cervical cancer is a novel treatment technique. The purpose of this study was to report our analysis of dose-volume parameters predicting gross tumor volume (GTV) control. Methods and Materials: We analyzed the records of 134 patients with International Federation of Gynecology and Obstetrics stages IB1-IVB cervical cancer treated with combined MRI-guided HDR and IMRT from July 2009 to July 2011. IMRT was targeted to the metabolic tumor volume and lymph nodes by use of FDG-PET/CT simulation. The GTV for each HDR fraction was delineated by use of T2-weighted or apparent diffusion coefficient maps from diffusion-weighted sequences. The D100, D90, and Dmean delivered to the GTV from HDR and IMRT were summed to EQD2. Results: One hundred twenty-five patients received all irradiation treatment as planned, and 9 did not complete treatment. All 134 patients are included in this analysis. Treatment failure in the cervix occurred in 24 patients (18.0%). Patients with cervix failures had a lower D100, D90, and Dmean than those who did not experience failure in the cervix. The respective doses to the GTV were 41, 58, and 136 Gy for failures compared with 67, 99, and 236 Gy for those who did not experience failure (P<.001). Probit analysis estimated the minimum D100, D90, and Dmean doses required for ≥90% local control to be 69, 98, and 260 Gy (P<.001). Conclusions: Total dose delivered to the GTV from combined MRI-guided HDR and PET/CT-guided IMRT is highly correlated with local tumor control. The findings can be directly applied in the clinic for dose adaptation to maximize local control.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Influence of Noncompliance With Radiation Therapy Protocol Guidelines and Operative Bed Recurrences for Children With Rhabdomyosarcoma and Microscopic Residual Disease: A Report From the Children's Oncology Group

Purpose: Postoperative radiation therapy (RT) is recommended for patients with rhabdomyosarcoma having microscopic disease. Sometimes RT dose/volume is reduced or omitted in an attempt to avoid late effects, particularly in young children. We reviewed operative bed recurrences to determine if noncompliance with RT protocol guidelines influenced local-regional control. Methods and Materials: All operative bed recurrences among 695 Group II rhabdomyosarcoma patients in Intergroup Rhabdomyosarcoma Study Group (IRS) I through IV were reviewed for deviation from RT protocol. Major/minor dose deviation was defined as >10% or 6-10% of the prescribed dose (40-60 Gy), respectively. Major/minor volume deviation was defined as tumor excluded from the RT field or treatment volume not covered by the specified margin (preoperative tumor volume and 2- to 5-cm margin), respectively. No RT was a major deviation. Results: Forty-six of 83 (55%) patients with operative bed recurrences did not receive the intended RT (39 major and 7 minor deviations). RT omission was the most frequent RT protocol deviation (19/46, 41%), followed by dose (17/46, 37%), volume (9/46, 20%), and dose and volume deviation (1/46, 2%). Only 7 operative bed recurrences occurred in IRS IV (5% local-regional failure) with only 3 RT protocol deviations. Sixty-three (76%) patients with recurrence died of disease despite retrieval therapy, including 13 of 19 nonirradiated children. Conclusion: Over half of the operative bed recurrences were associated with noncompliance; omission of RT was the most common protocol deviation. Three fourths of children die when local-regional disease is not controlled, emphasizing the importance of RT in Group II rhabdomyosarcoma.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Ten questions concerning reinforcement learning for building energy management

As buildings account for approximately 40% of global energy consumption and associated greenhouse gas emissions, their role in decarbonizing the power grid is crucial. The increased integration of variable energy sources, such as renewables, introduces uncertainties and unprecedented flexibilities, necessitating buildings to adapt their energy demand to enhance grid resiliency. Consequently, buildings must transition from passive energy consumers to active grid assets, providing demand flexibility and energy elasticity while maintaining occupant comfort and health. This fundamental shift demands advanced optimal control methods to manage escalating energy demand and avert power outages. Reinforcement learning (RL) emerges as a promising method to address these challenges. Here, in this paper, we explore ten questions related to the application of RL in buildings, specifically targeting flexible energy management. We consider the growing availability of data, advancements in machine learning algorithms, open-source tools, and the practical deployment aspects associated with software and hardware requirements. Our objective is to deliver a comprehensive introduction to RL, present an overview of existing research and accomplishments, underscore the challenges and opportunities, and propose potential future research directions to expedite the adoption of RL for building energy management.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Nonresected Non-Small-Cell Lung Cancer in Stages I Through IIIB: Accelerated, Twice-Daily, High-Dose Radiotherapy-A Prospective Phase I/II Trial With Long-Term Follow-Up

Purpose: Our purpose was to investigate the tolerability of accelerated, twice-daily, high-dose radiotherapy. The secondary endpoints were survival and locoregional tumor control. Methods and Materials: Thirty consecutive patients with histologically/cytologically proven non-small-cell lung cancer were enrolled. Tumor Stage I, II, IIIA, and IIIB was found in 7, 3, 12, and 8 patients, respectively. We applied a median of 84.6 Gy (range, 75.6-90.0 Gy) to the primary tumors, 63.0 Gy (range, 59.4-72.0 Gy) to lymph nodes, and 45 Gy to nodes electively (within a region of about 6 cm cranial to macroscopically involved sites). Fractional doses of 1.8 Gy twice daily, with an interval of 11 hours, were given, resulting in a median treatment time of 35 days. In the majority of patients the conformal target-splitting technique was used. In 19 patients (63%) two cycles of induction chemotherapy were given. The median follow-up time of survivors is 72 months (range, 62-74 months). Results: We found Grade 1, 2 and 3 acute esophageal toxicity in 11 patients (37%), 2 patients (7%), and 2 patients (7%), respectively. Grade 2 acute pneumonitis was seen in 2 patients (7%). No late toxicity greater than Grade 1 was observed. The actual overall survival rates at 2 and 5 years are 63% and 23%, respectively; the median overall survival, 27.7 months. In 9 patients a local failure occurred, 7 of them presenting initially with an atelectasis without availability of 18-fluorodeoxyglucose-positron emission tomography staging at that time. In 4 patients recurrence occurred regionally. Conclusions: This Phase I/II trial with long-term follow-up shows low toxicity with promising results for survival and locoregional tumor control.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Controlled synthesis and gas sensing properties of In{sub 2}O{sub 3} with different phases from urchin-like InOOH microspheres

Highlights: • Preparation of nanostructured In{sub 2}O{sub 3} microspheres. • Morphology and phase control of In{sub 2}O{sub 3}. • Gas sensors based on the In{sub 2}O{sub 3} microspheres exhibit excellent sensing properties for the detection of formaldehyde. - Abstract: Urchin-like InOOH microspheres were successfully prepared by a convenient and controllable method. Such experimental parameters as solvents and complexing reagents on the formation of the urchin-like InOOH microspheres were investigated. Scanning electron microscopy, X-ray diffraction and infrared spectroscopy were employed to investigate the evolution process of the urchin-like InOOH precursors. Furthermore, the formation mechanism of the urchin-like InOOH microspheres was proposed. By annealing the urchin-like InOOH precursor at different temperatures under ambient pressure, rhombohedral corundum-type indium oxide (rh-In{sub 2}O{sub 3}), cubic bixbyite-type indium oxide (c-In{sub 2}O{sub 3}) and mixed phases of rh-In{sub 2}O{sub 3} and c-In{sub 2}O{sub 3} were obtained. The gas sensing properties of the prepared In{sub 2}O{sub 3} samples were examined. It was found that the sensors based on the prepared In{sub 2}O{sub 3} samples exhibited excellent response and selectivity to formaldehyde.

36 MATERIALS SCIENCE↗

First season MWA EoR power spectrum results at redshift 7

The Murchison Widefield Array (MWA) has collected hundreds of hours of Epoch of Reionization (EoR) data and now faces the challenge of overcoming foreground and systematic contamination to reduce the data to a cosmological measurement. We introduce several novel analysis techniques, such as cable reflection calibration, hyper-resolution gridding kernels, diffuse foreground model subtraction, and quality control methods. Each change to the analysis pipeline is tested against a two-dimensional power spectrum figure of merit to demonstrate improvement. We incorporate the new techniques into a deep integration of 32 hours of MWA data. This data set is used to place a systematic-limited upper limit on the cosmological power spectrum of Δ{sup 2}⩽2.7×10{sup 4} mK{sup 2} at k = 0.27 h Mpc{sup −1} and z = 7.1, consistent with other published limits, and a modest improvement (factor of 1.4) over previous MWA results. From this deep analysis, we have identified a list of improvements to be made to our EoR data analysis strategies. These improvements will be implemented in the future and detailed in upcoming publications.

79 ASTRONOMY AND ASTROPHYSICS↗

Grey-box and ANN-based building models for multistep-ahead prediction of indoor temperature to implement model predictive control

Model-based predictive control (MPC) strategies for heating, ventilation, and air-conditioning (HVAC) systems present an opportunity to lower building energy consumption and operational costs. Such approaches rely on the development of a model to precisely forecast building thermal dynamics, such as room air temperature or heating/cooling rate, and make control-related decisions. The control-oriented modeling of building energy systems should be accurate in predicting indoor conditions and present low computational complexity. These features are the key challenge of implementing advanced control methods such as MPC. Extant studies on building modeling for MPC have focused on step-ahead forecasting techniques to forecast building thermal dynamics, while multistep-ahead forecasting is essential. Moreover, machine learning model suitable in case of the domain-based engineering expertise are also not available. To this aim, we perform a comparative analysis of the grey-box model based on a resistance-capacitance (RC) thermal network and a machine learning model composed of an artificial neural network (ANN) for multistep-ahead prediction of building thermal dynamics using current and historical data. Actual experimental data obtained from the Flexible Research Platform (FRP) in Oak Ridge National Laboratory (US) are used for estimation and validation purposes. The average root mean squared error (RMSE) of the grey-box and ANN models are 0.89 °C and 1.02°C, respectively. Finally, the results indicate that the grey-box model outperforms the ANN model in the considered validation periods in terms of accuracy and prediction stability.

42 ENGINEERING↗

Relating Dose Outside the Prostate With Freedom From Failure in the Dutch Trial 68 Gy vs. 78 Gy

Purpose: For prostate cancer patients at risk for subclinical spread of the disease, we investigated whether incidental dose outside the target was associated with tumor control. Methods and Materials: We selected 352 intermediate-risk (mainly T2b-T3a) and high-risk (mainly T3b) patients treated in a randomized trial. Target volume was prostate (68-78 Gy) and seminal vesicles (50-78 Gy). Failure (clinical or biochemical) was evaluated at 4 years. To compare three-dimensional dose distributions, an automated mapping procedure was introduced. Between patients, these maps provide an approximate identification of corresponding anatomical locations. Maps of the dose difference between patients with and without failure were constructed. Univariate and multivariate analyses were performed including the dose in selected points. Results: Dose differences were mainly found in the obturatorial region for the high-risk patients, and in the presacral region for the intermediate risk group (>7 Gy, p < 0.01). Univariate hazard ratios per 10 Gy for selected dose points were 0.83 (p = 0.01, obturatorial) and 0.72 (p = 0.002, presacral). These hazard ratios were stable under multivariate analysis correcting for established prognostic factors, hospital, and dose to the prostate. Conclusions: Patients without failure have received on average a higher dose to regions where regional cancer spread could be expected.

62 RADIOLOGY AND NUCLEAR MEDICINE↗