Search

Abstract Background 90% of glioblastomas (GBM) relapse within two years of diagnosis. In contrast to the initial setting, there is no standard management for recurrent disease and options include hypofractionated stereotactic re-irradiation (re-mHSRT). The aims of this study were to investigate re-mHSRT practice in Swiss neuro-oncology centres. Methods A survey of 18 questions regarding re-irradiation for GBM was created and distributed electronically (SurveyMonkey, USA) to 11 radiation oncologists in Switzerland specialising in brain tumours. We evaluated the clinical outcomes of a multicentre series of patients treated with an established re-mHSRT schedule to benchmark these against the literature and investigated the radiological patterns of relapse after re-mHSRT. Results 8 of 11 (73%) radiation oncologists responded to the survey and re-irradiation practice was heterogeneous. The 10 × 3.5 Gy schedule (RTOG 1205, BRIOChe trials) was used by 5/8 respondents and 47/50 patients with recurrent GBM treated with re-mHSRT with this schedule in daily practice were included in the analysis. The median time to re-mHSRT following completion of adjuvant RT was 23.3 (7-224) months. The median PTV at re-mHSRT was 22.0 cm3 (0.9–190). Combined CTV + PTV margins ranged from 0 to 10 mm and median prescription isodose was 80% (67–100). 14/47 (30%) patients received temozolomide and four (8.5%) continued bevacizumab concomitantly. On multivariable analysis, concomitant systemic therapy predicted for progression-free survival (PFS), HR 2.87 (95% CI 1-03-7.96), p = 0.042. Median PFS following re-mHSRT was 6.6 (0.2–92.5) months and 26/47 patients (55%) received subsequent systemic therapy. The median overall survival (OS) following recurrence was 11.8 months (1.5–92.5), similar to the 10.8 months in the literature with the same schedule. The six-month OS rate was 37/47 (79%), which compares well with the 73% reported in a meta-analysis of 50 publications employing various schedules. In a subgroup analysis of 36/47 (79%) patients with MR follow-up after re-mHSRT, 8/36 (22%) had no radiological evidence of tumour progression at a median follow-up of 9.4 months. 21/28 (75%) radiological relapses were in-field, two were marginal and five were out of field. Conclusions Re-mHSRT with 10 × 3.5 Gy can achieve local control in selected patients with recurrent GBM.

Abstract This research provides new insights into the composition of zinc white paints used by Piet Mondrian during his neoplastic period. Nine paintings, dated between 1921 and 1935, were studied, with a focus on three works in the collection of the Fondation Beyeler (Basel, Switzerland)—Tableau I (1921–1925), Composition with yellow and blue (1932) and Composition with double line and blue (1935)—and on Lozenge composition with yellow lines (1933) in Kunstmuseum Den Haag (The Netherlands). Cross sections from other paintings, most of which previously studied by Van Asperen de Boer in the early 1990’s, were reexamined as well. The analyses revealed a zinc white paint with aluminum phosphate inclusions (ZW-Al/P). In two of the works, a zinc white paint with aluminum sulphate was also found. The occurrence of aluminum phosphate (or aluminum sulphate) in 1920–1930’s paint formulations, and in paintings by Mondrian, has never been reported so far. Likely, the use of this zinc white paint in Mondrian’s and other artworks is more widespread than currently known, but it may have been overlooked in similar case studies, since its identification can only be accomplished with detailed scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM–EDX) analysis of cross sections. Mondrian’s use of pure ‘bright’ ZW-Al/P paint is limited to the later paintings (1932–1935) of this study, whereas in the earlier works (1921–1929) it was mostly mixed with lead white, likely conveying a different hue. Both the ZW-Al/P and Zn-Al/S ‘marker paints’ were also added to yellow, blue or black paint in specific paint layers of the planes and lines, respectively. It was also shown that Mondrian used the same paint on the frames as in the white planes and that the frames were painted while still working on the composition, and not only when he considered the work completed. Indeed, the detection of the zinc white marker paints might help to delineate the chronology of Mondrian’s working process in other paintings and may also be important in the interpretation of overpaints in the course of a treatment when removal of certain paint layers is considered.

We investigate the stationary (late-time) training regime of single- and two-layer underparameterized linear neural networks within the continuum limit of stochastic gradient descent (SGD) for synthetic Gaussian data. In the case of a single-layer network in the weakly underparameterized regime, the spectrum of the noise covariance matrix deviates notably from the Hessian, which can be attributed to the broken detailed balance of SGD dynamics. The weight fluctuations, are in this case, generally anisotropic but effectively experience an isotropic loss. For an underparameterized two-layer network, we describe the stochastic dynamics of the weights in each layer and analyze the associated stationary covariances. We identify the interlayer coupling as a distinct source of anisotropy for the weight fluctuations. In contrast to the single-layer case, the weight fluctuations are effectively subject to an anisotropic loss, the flatness of which is inversely related to the fluctuation variance. We thereby provide an analytical derivation of the recently observed inverse-variance flatness relation in a model of a deep linear neural network.

Abstract Many wind energy projects start with resource assessments based on outputs from numerical models followed by meteorological campaigns. These models are imperfect and have a substantial input parameter space. It is difficult to discern how the inputs affect the simulation results. Surface roughness is generally crudely represented in mesoscale models. In this work, we use the Weather Research and Forecast model to simulate winds at a meteorological mast in Mexico with wind observations at 80 m. The model sensitivity to changes in surface roughness is contrasted with permissively perturbed orography fields to gauge the observed changes' relative importance. Changes in surface roughness affect the root mean square error of the simulated 80‐m winds compared to the observations, in the same order of magnitude as orography perturbations. Using a roughness field derived from a synthetic aperture radar improved the wind speed bias. Wind speed predictions using high‐resolution (185 m) simulations sometimes showed excellent agreement with the observations, but there were several instances when medium‐resolution (1.6 km) simulations performed better. A wind speed time series with high temporal (10 min) resolution from a low spatial (6 km) resolution simulation was used to train a deep neural network regression (DNNR) model. The trained DNNR reduced the wind speed error the most compared to any of the other simulations performed in this study. The second‐best model performance was obtained using the radar roughness‐derived field. The results of the simulations at medium resolution with perturbed orography were very similar to those using different roughness fields inputs.

Convective ensembles promise to increase forecast accuracy while at the same time providing information on the probability of the forecast. A vast number of different methods of ensemble creation have been developed over time. Here, initial conditions and model error uncertainties are represented by a convective-allowing ensemble with more than 50 members. The results are analyzed using one case study with relatively high precipitation over Ensenada, Baja California, Mexico. The ensemble members are perturbed using random initial perturbations, breeding, and the Stochastic Kinetic Energy Backscatter parameterization (SKEBS) within the Weather Research and Forecasting (WRF) model. The aim is to improve the high-resolution ensemble design provided in a previous study for the same region by maximizing the spread of an ensemble with low member count. To this end, a comparative analysis of the members is performed using perturbation growth rates and information entropy. In addition, a comparative verification is performed using observations from one automatic meteorological station and satellite-derived precipitation data. It was found that the growth rates and the one-dimensional power spectral density of the initial perturbation fields are clustered depending on each member’s origin and the methods used to generate the breeding members. An inverse relationship was observed between these two variables, which can be useful for selecting appropriate initial condition perturbations. The dynamical injections of energy, introduced as perturbations to the numerical fields by the SKEBS method, were essential to maintain positive growth rates during the simulation period. Evaluation of the information entropy suggests that a selection of a set of members generated by the SKEBS method is best for increasing the ensemble spread while saving computer resources.

Tidal renewable energy is a promising alternative energy resource, and marginal seas are known as excellent sites for tidal energy exploitation. In-stream and tidal energy devices are less exposed to extreme weather events than wind energy. Nevertheless, during tropical storms, the currents may intensify to levels that threaten the integrity of the devices. This paper presents Hurricane Odile and its impact on the currents in the Gulf of California (GC) as a worst case scenario. A methodology to analyze the impact and its potential effects on tidal energy converters installed within the region are presented. The analysis is based on predictions obtained with a 3D shallow water model forced by tides and the meteorological conditions generated by Odile. A tidal model with no wind forcing was used for validation of the tidal model predictions. After validation, the two models were used to analyze the maximum anomaly in surface currents and sea surface height caused by the passage of Odile, and to analyze at which depth the devices could be deemed safe from any impact of the hurricane. Some anomalies extended throughout the water column, even in the deep regions of the GC. This paper highlights the importance of including the meteorological forcing in evaluations of tidal range or in-stream renewable energy resources and introduces new measures of device exposure to the current anomalies.

In previous work, the authors demonstrated how data from climate simulations can be utilized to estimate regional wind power densities. In particular, it was shown that the quality of wind power densities, estimated from the UPSCALE global dataset in offshore regions of Mexico, compared well with regional high resolution studies. Additionally, a link between surface temperature and moist air density in the estimates was presented. UPSCALE is an acronym for UK on PRACE (the Partnership for Advanced Computing in Europe)-weather-resolving Simulations of Climate for globAL Environmental risk. The UPSCALE experiment was performed in 2012 by NCAS (National Centre for Atmospheric Science)-Climate, at the University of Reading and the UK Met Office Hadley Centre. The study included a 25.6-year, five-member ensemble simulation of the HadGEM3 global atmosphere, at 25km resolution for present climate conditions. The initial conditions for the ensemble runs were taken from consecutive days of a test configuration. In the present paper, the emphasis is placed on the single climate run for a potential future climate scenario in the UPSCALE experiment dataset, using the Representation Concentrations Pathways (RCP) 8.5 climate change scenario. Firstly, some tests were performed to ensure that the results using only one instantiation of the current climate dataset are as robust as possible within the constraints of the available data. In order to achieve this, an artificial time series over a longer sampling period was created. Then, it was shown that these longer time series provided almost the same results than the short ones, thus leading to the argument that the short time series is sufficient to capture the climate. Finally, with the confidence that one instantiation is sufficient, the future climate dataset was analysed to provide, for the first time, a projection of future changes in wind power resources using the UPSCALE dataset. It is hoped that this, in turn, will provide some guidance for wind power developers and policy makers to prepare and adapt for climate change impacts on wind energy production. Although offshore locations around Mexico were used as a case study, the dataset is global and hence the methodology presented can be readily applied at any desired location.