Impact Of Random Errors Research Articles

In-Motion Coarse Alignment Method Based on Position Loci and Optimal-REQUEST for SINS

In this paper, an improved in-motion coarse alignment method is proposed for a strapdown inertial navigation system (SINS) using position loci obtained from the Global Positioning System (GPS). The difference from the popular coarse alignment methods is that the proposed algorithm uses GPS position loci information to form the vector observation, and does not need velocity information, which expands the application range of in-motion coarse alignment. In addition, this paper utilizes the Optimal-REQUEST algorithm to reduce the influence of random errors contained in the vector observation. The Optimal-REQUEST algorithm is an adaptive iterative updating algorithm, which can adaptively adjust the gain of the filter according to the loss function. Simulation results confirmed that the proposed algorithm can suppress the impact of random errors effectively. The pitch, roll and yaw angles calculated by the proposed algorithm were improved by 51.95%, 53.80% and 63.03% compared with the comparison algorithms.

Toward the Removal of Model Dependency in Soil Moisture Climate Data Records by Using an $L$-Band Scaling Reference

Building climate data records of soil moisture (SM) requires computing long time series by merging retrievals from sensors on-board different satellites, which implies to perform a bias correction or rescaling on the original time series. Due to their long time span and high temporal frequency, model data could be used as a common reference for the rescaling. However, avoiding model dependence in observational climate data records is needed for some applications. In this article, the possibility of using as reference remote sensing data from one of the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$L$</tex-math></inline-formula> -band sensors specifically designed to measure SM is discussed. Advanced Microwave Scanning Radiometer 2 SM time series were rescaled by matching their cumulative distribution functions (CDFs) to those of Soil Moisture and Ocean Salinity (SMOS), Soil Moisture Active Passive (SMAP), and Global Land Data Assimilation System (GLDAS) NOAH model time series. The CDF computation was investigated as a function of the time series length, finding significant differences from four to nine years. Replacing temporal by spatial variance does not allow us to compute better CDFs from short time series. The rescaled time series show a high correlation ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$R&gt;0.8$</tex-math></inline-formula> ) to the original ones and a low bias with respect to the reference ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$&lt; $</tex-math></inline-formula> 0.03 m <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^{3}\cdot$</tex-math></inline-formula> m <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^{-3}$</tex-math></inline-formula> ). The time series rescaled using several SMOS or SMAP datasets were also evaluated against <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in situ</i> measurements and show performances similar to or slightly better than those rescaled using the model GLDAS. The impact of random errors and gaps of the observational data into the rescaling was evaluated. These results show that it is actually possible to use <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$L$</tex-math></inline-formula> -band data as reference to rescale time series from other sensors to build long time series of SM.

Fitting the Search Space of Weight-sharing NAS with Graph Convolutional Networks

Neural architecture search has attracted wide attentions in both academia and industry. To accelerate it, researchers proposed weight-sharing methods which first train a super-network to reuse computation among different operators, from which exponentially many sub-networks can be sampled and efficiently evaluated. These methods enjoy great advantages in terms of computational costs, but the sampled sub-networks are not guaranteed to be estimated precisely unless an individual training process is taken. This paper owes such inaccuracy to the inevitable mismatch between assembled network layers, so that there is a random error term added to each estimation. We alleviate this issue by training a graph convolutional network to fit the performance of sampled sub-networks so that the impact of random errors becomes minimal. With this strategy, we achieve a higher rank correlation coefficient in the selected set of candidates, which consequently leads to better performance of the final architecture. In addition, our approach also enjoys the flexibility of being used under different hardware constraints, since the graph convolutional network has provided an efficient lookup table of the performance of architectures in the entire search space.

Evaluating the dosimetric consequences of MLC leaf positioning errors in dynamic IMRT treatments

AbstractPurposeThe purpose of this study was to evaluate the dosimetric impact of multileaf collimator (MLC) positional errors on dynamic intensity-modulated radiotherapy (IMRT) treatments through planning simulation. Secondly the sensitivity of IMRT MatriXX device for detecting the MLC leaf positional errors was also evaluated.Materials and methodsIn this study five dynamic IMRT plans, each for brain and head–neck (HN), were retrospectively included. An in-house software was used to introduce random errors (uniform distribution between −2·0 and +2·0 mm) and systematic errors [±0·5, ±0·75, ±1·0 and ±2·0 mm (+: open MLC error and −: close MLC error)]. The error-introduced MLC files were imported into the treatment planning system and new dose distributions were calculated. Furthermore, the dose–volume histogram files of all plans were exported to in-house software for equivalent uniform dose (EUD), tumour control probability and normal tissue complication probability calculations. The error-introduced plans were also delivered on LINAC, and the planar fluences were measured by IMRT MatriXX. Further, 3%/3 mm and 2%/2 mm γ-criteria were used for analysis.ResultsIn planning simulation study, the impact of random errors was negligible and ΔEUD was &amp;lt;0·5±0·7%, for both brain and HN. The impact of systematic errors was substantial, and on average, the maximum change in EUD for systematic errors (close 2 mm) was −10·7±3·1% for brain and −15·5±2·6% for HN.ConclusionsIt can be concluded that the acceptable systematic error was 0·4 mm for brain and 0·3 mm for HN. Furthermore, IMRT MatriXX device was able to detect the MLC errors ≥2 mm in HN and &amp;gt;3 mm errors in brain with 2%/2 mm γ-criteria.

Evaluating the Consequences of MLC Leaf Positioning Errors in Dynamic IMRT Treatments

MLC positioning accuracy in dynamic IMRT is very important for precise treatment delivery and any mismatch between planned and actual treatment delivery can lead to serious dose discrepancies. The purpose of this simulation study was to evaluate the dosimetric impact of leaf positional errors in dynamic IMRT treatments. Dynamic IMRT plans of five head-neck (HN) (66Gy/30# for CTV1 and 54Gy/30# for CTV2) and five brain (54Gy/30#) patients were retrospectively included in this study. MLC files of these plans were exported to in-house developed software that introduced known leaf positional errors. Two types of errors were considered; random errors (uniform distribution between -2.0 to +2.0mm) and systematic errors (±0.5, ±0.75, ±1.0 & ±2.0 mm. The error introduced MLC files were imported into the TPS and dose distributions were re-calculated. The error plans were compared against the respective original plans. The DVH files were exported to in-house software for TCP and NTCP calculation. The impact of random errors was observed to be negligible (<0.5%) while that of systematic errors was substantial. On average, the changes in TCP and NTCP for all systematic errors were less than 5% for brain. On the other hand, the effect was observed to be more prominent in HN. Compared to reference plans the change in TCP of CTV1 was 4% for 0.5mm and 7.0-26.5% for 0.75mm to 2.0mm; whereas, the same for CTV2 was <1.0% up to 1.0mm and 5.1% for 2.0mm error. Similarly, the changes in NTCP of left and right parotid were 10.8% (2.5Gy difference in mean dose) and 4.5% (1.8Gy difference in mean dose) for 0.5mm error respectively; 12.7%-23.1% (2.7-7.0Gy difference in mean dose) and 5.5%-12.8% (2.5-6.3Gy difference in mean dose) for 0.75mm to 2.0mm.Abstract 3657; TableTCP / NTCP (Mean ± SD) in Head and Neck casesPlanCTV1CTV2PTV1PTV2Lt ParotidRT ParotidRef88.1±0.898.8±0.488.0±0.798.7±0.514.1±22.94.7±10.5Random88.4±0.498.8±0.488.2±0.398.7±0.415.2±24.74.6±10.2+0.5 mm91.7±0.899.2±0.391.6±0.599.1±0.323.3±35.08.2±18.3+0.75 mm92.8±0.999.3±0.292.8±0.699.3±0.226.8±39.010.2±22.8+1 mm93.9±0.999.4±0.293.9±0.799.4±0.229.9±42.412.1±27.1+2 mm96.6±0.999.7±0.196.6±0.799.7±0.137.2±51.017.5±39.1-0.5 mm84.4±0.898.3±0.684.0±1.098.1±0.78.3±13.92.4±5.3-0.75 mm81.7±1.397.9±0.781.2±1.697.7±0.85.5±9.31.6±3.6-1 mm78.6±2.097.5±0.977.9±2.497.1±1.13.5±5.91.1±2.4-2mm61.6±6.693.7±2.459.5±7.092.4±3.20.4±0.70.2+0.4 Open table in a new tab Systematic errors of MLC in dynamic IMRT showed considerable biological consequences whereas the random error resulted in negligible effect. Based on this study, it can be concluded that the acceptable systematic error is up to1.0mm for brain and up to 0.5mm for HN.

SU‐E‐T‐789: A Study on the Sensitivity of VMAT and IMRT Prostate Plans Considering Uncertainties in Treatment Delivery and Patient Positioning

Purpose: During the course of radiotherapy random uncertainties in patient positioning and the fluctuation of machine parameters are inevitable. In this study we compared the impact of these errors on the quality of delivered Volumetric Modulated Arc Therapy and Intensity Modulated Radiation Therapy prostate plans. Methods: VMAT and IMRT plans were generated for two representative contour sets in Pinnacle. The uncertainty in patient setup and machine parameters during delivery was simulated by introducing random errors in isocenter position, MLC leaf positions and gantry angles for each fraction of the treatment. Random error ranges of ±3mm, ±1mm and ±1deg were used for isocenter position, MLC leaf positions and gantry angles respectively. The simulated plans were recalculated in Pinnacle and the impact of random errors on the plan quality was studied by comparing the Dose Volume Histogram of target volumes and OARs for both VMAT and IMRT plans. Results: The introduction of random errors resulted in dose smearing of both VMAT and IMRT plans. Less than 1% difference in the dose to the CTV was seen between the planning and error simulated plans for both VMAT and IMRT. However the difference in dose between the planned and error simulated plans for the PTV was greater for VMAT compared with IMRT. TheV95 of PTV was reduced by less than 1% for IMRT, whereas a difference of up to 2% was observed in VMAT. A variation of less than 1% in EUD for both the rectum and bladder was observed between the planned and error plans for both VMAT and IMRT. Conclusions: VMAT plans are relatively more sensitive to random errors in treatment delivery and patient positioning compared to IMRT plans when considering dose to the PTV, however only small variations are seen in both the CTV and OARs for both VMAT and IMRT plans.

Poster — Thur Eve — 61: Intensity Modulated Radiation Therapy: The Relationship between Planar Dose Map Verification and Dosimetric Outcome

This work analyses the consequences of random and systematic dose differences in the comparison of calculated against measured planar dose distributions for IMRT from the perspective of a surrogate of the treatment outcome: Equivalent Uniform Dose (EUD). In‐house software was developed to incorporate normally distributed errors in the fluence maps of 3 head and neck (H&amp;N) and 3 prostate plans and simulate dose differences that appear randomly across the planar dose maps. The plans with random errors were grouped according to the following passing rates: 1) 90–95%, 2)85–90% and 3)80–85% during patient specific quality control. The passing criteria included a 3% absolute dose, 3 mm distance to agreement (DTA), and a minimum dose difference of 2 cGy. A systematic 1% dose error could also be incorporated by altering the MUs of each plan with random errors. The impact of random errors on the prescribed EUDs of H&amp;N plans ranged from −2.7 to −1.3% for the CTV and −1.0 to 0.6 Gy for the OARs while in prostate plans they ranged from −1.6 to −0.6% for the CTV and −1.2 to −0.4 Gy for the OARs over the range examined. The criterion of 90% passing rate for 3% absolute dose and 3 mm DTA kept the effects of random errors within a dosimetric goal of 2% change in prescribed EUDs of the targets and 2 Gy for the OARs. Systematic errors, if present, may cause larger effects on clinical dosimetry while still meeting patient specific quality control tolerances.

Reflector Antenna Distortion Analysis Using MEFCM

The impact of random errors and systematic errors due to reflector surface distortions is discussed by using Ruze formula in conjunction with the mechanical-electromagnetic-field coupling model (MEFCM). Ruze formula demands too rigorous reflector precision and manufacturing accuracy due to not taking the systematic error into consideration. The best-fitting method treats the systematic error, but not random error. The MEFCM takes both systematic error and random error into account. Numerical examples of 7.3 m and 40 m diameter reflector antenna validate the method proposed. The proposed MEFCM method integrates the mechanical design with electromagnetic analysis in order to optimize the reflector antenna design.

EVALUATION OF THE INFLUENCE OF UNCERTAINTY IN RAINFALL AND DISCHARGE DATA ON HYDROLOGICAL MODELING

The objectives of this study are to analyze the influence of systematic and random error inrainfall data, discharge data, and the spatial representation of rainfall data on the performance of a distributed hydrologicalmodel, BTOPMC. A framework for uncertainty analysis was developed using a Monte Carlo approach, which was applied to the Kalu River basin in Sri Lanka. Findings show that a systematic error exceeding +/-10% in rainfall or discharge data is detrimental to model results. A random error with standarddeviation is equal to10% ofrainfall or discharge isnot substantial. Calibrationof parameters can compensate for some error. The impact of systematic error in rainfall in terms of Nash-Sutcliffe Efficiency (NSE) is higher than that in discharge. The impact of random error in discharge in terms of NSE is higher than that inrainfall. The impact of the random error is the lowest for the gauge network of the highest density, but the impact of systematic error is not the lowest for this case.

The Effects of Calcium Channel Blockers on Cardiovascular Outcomes: a Review of Randomised Controlled Trials

The choice of antihypertensive treatment should be guided by evidence of a reduction in the risk of cardiovascular (CV) events and therefore improved long-term outcome. Using pre-determined criteria, ten randomised, controlled trials that assessed the effects of calcium channel blockers (CCBs) on CV events in patients with hypertension were identified. Six of them enrolled a relatively small number (<1500) of hypertensive patients, whereas four of the studies were much larger (>4500 patients). The smaller studies produced mixed findings, especially those trials where CCBs were compared with diuretics; this may reflect methodological limitations and the impact of random error. The results from the four larger studies produced a consistent message: long-acting CCBs such as nifedipine, administered in a gastro-intestinal-transport-system (GITS) formulation, nitrendipine and diltiazem, reduce CV morbidity and mortality in hypertensive patients. In the one study that specifically enrolled high-risk hypertensive patients, nifedipine GITS was as effective as diuretic therapy in reducing CV events, and in all four larger studies sub-group analyses showed that the benefits of these CCBs extend to hypertensive patients with diabetes. The available evidence supports the use of these long-acting CCBs as a first-line treatment option in hypertensive patients.

DATUM DEFINITION FOR GPS NETWORKS

This paper focuses on various aspects of datum definitions and datum design for GPS networks. We consider GPS observations as being a major source for positioning future control points. Hence accuracy, reliability, and sensitivity of the datum definition become important.Accuracy measures the impact of random errors, and reliability measures the impact of gross errors in coordinate estimation. Sensitivity provides a quality measure for how well point movements are estimated in time.It turns out that the accuracy and the reliability of GPS networks are independent of the geometric distribution of the network points; this is in contrast to control networks measured by classical methods. We demonstrate that the distribution of datum points does not affect accuracy and reliability while sensitivity does depend on this distribution. We present a number of rules for the distribution of datum points that improve the sensitivity of the datum.The internal reliability does not depend on datum definition. The external reliability, the accuracy and the sensitivity depend on datum definition. It turns out that Baarda's suggestion for measuring the global external reliability is not suitable for GPS networks, another procedure is given.