Search

The spread of high-performance personal computers, frequently equipped with powerful Graphic Processing Units (GPUs), has raised interest in a set of techniques that are able to extract models of electromagnetic phenomena (and devices) directly from available examples of desired behavior. Such approaches are collectively referred to as Machine Learning (ML). A typical representative ML approach is the so-called "Neural Network" (NN). Using such data-driven models allows the evaluation of the output in a much shorter time when a theoretical model is available, or allows the prediction of the behavior of the systems and devices when no theoretical model is available. With reference to a simple yet representative benchmark electromagnetic problem, some of the possibilities and pitfalls of the use of NNs for the interpretation of measurements (inverse problem) or to obtain required measurements (optimal design problem) are discussed. The investigated aspects include the choice of NN model, the generation of the dataset(s), and the selection of hyper-parameters (hidden layers, training paradigm). Finally, the capabilities in the handling of ill-posed problems are critically revised. [ABSTRACT FROM AUTHOR]

Background: In ARDS caused by COVID-19 pneumonia, appropriate adjustment of physiologic parameters based on lung stretch or oxygenation may optimize the ventilatory strategy. This study aims to describe the prognostic performance on 60-d mortality of single and composite respiratory variables in subjects with COVID-19 ARDS who are on mechanical ventilation with a lung-protective strategy, including the oxygenation stretch index combining oxygenation and driving pressure (ΔP). Methods: This single-center observational cohort study enrolled 166 subjects on mechanical ventilation and diagnosed with COVID-19 ARDS. We evaluated their clinical and physiologic characteristics. The primary study outcome was 60-d mortality. Prognostic factors were evaluated through receiver operating characteristic analysis, Cox proportional hazards regression model, and Kaplan-Meier survival curves. Results: Mortality at day 60 was 18.1%, and hospital mortality was 22.9%. Oxygenation, ΔP, and composite variables were tested: oxygenation stretch index ([Formula: see text]/[Formula: see text] divided by ΔP) and ΔP x 4 + breathing frequency (f) (ΔP x 4 + f). At both day 1 and day 2 after inclusion, the oxygenation stretch index had the best area under the receiver operating characteristic curve (oxygenation stretch index on day 1 0.76 (95% CI 0.67-0.84) and on day 2 0.83 (95% CI 0.76-0.91) to predict 60-d mortality, although without significant difference from other indexes. In multivariable Cox regression, ΔP, [Formula: see text]/[Formula: see text], ΔP x 4 + f, and oxygenation stretch index were all associated with 60-d mortality. When dichotomizing the variables, ΔP ≥ 14, [Formula: see text]/[Formula: see text] ≤ 152 mm Hg, ΔP x 4 + f ≥ 80, and oxygenation stretch index < 7.7 showed lower 60-d survival probability. At day 2, after optimization of ventilatory settings, the subjects who persisted with the worse cutoff values for the oxygenation stretch index showed a lower probability of survival at 60 d compared with day 1; this was not the case for other parameters. Conclusions: The oxygenation stretch index, which combines [Formula: see text]/[Formula: see text] and ΔP, is associated with mortality and may be useful to predict clinical outcomes in COVID-19 ARDS. [ABSTRACT FROM AUTHOR]

In this contribution a methodology to diagnose transformer faults based on Dissolved Gas Analysis (DGA) by using a convolutional neural network (CNN) is proposed. The algorithm to transform the gas contents (resulting from the DGA analysis) into feature maps is introduced, and the resulting feature maps are the input of the CNN. In order to take into account the fact that the data set is imbalanced, the improved Synthetic Minority Over-Sampling Technique (SMOTE) is combined with the data cleaning technique to protect the CNN from training bias. The effect of the CNN architecture on the classification performance is also investigated to determine the optimal CNN parameters. All the above mentioned possibilities are tested and their performance investigated; in addition, a final test on the IEC TC 10 transformer fault database validates the accuracy and the generalization potential of the proposed methodology. [ABSTRACT FROM AUTHOR]

In this work, a new hydroelectric basin modelling approach is described and applied to the Pontecosi basin, Italy. Several types of data sources were used to learn the model: a number of weather stations, satellite observations, the reanalysis dataset, and basin data. With the goal of predicting the water level of the basin, the model was composed by three cascade modules. Firstly, different spatial interpolation methods, such as Kriging, Radial Basis Function, and Natural Neighbours, were compared and applied to interpolate the weather stations data nearby the basin area to infer the main environmental variables (air temperature, air humidity, precipitation, and wind speed) in the basin area. Then, using these variables as inputs, a neural network was trained to predict the mean soil moisture concentration over the area, also to improve the low availability due to satellite orbits. Finally, a non-linear auto regressive exogenous input (NARX) model was trained to simulate the basin level with different prediction horizons, using the data from the previous modules and past basin data (water level, discharge flow rate, and turbine flow rate). Accurate predictions of the basin water level were achieved within 1 to 6 h ahead, with mean absolute errors (MAE) between 2 cm and 10 cm, respectively. [ABSTRACT FROM AUTHOR]

Information about respiratory mechanics such as resistance, elastance, and muscular pressure is important to mitigate ventilator‐induced lung injury. Particularly during pressure support ventilation, the available options to quantify breathing effort and calculate respiratory system mechanics are often invasive or complex. We herein propose a robust and flexible estimation of respiratory effort better than current methods. We developed a method for non‐invasively estimating breathing effort using only flow and pressure signals. Mixed‐integer quadratic programming (MIQP) was employed, and the binary variables were the switching moments of the respiratory effort waveform. Mathematical constraints, based on ventilation physiology, were set for some variables to restrict feasible solutions. Simulated and patient data were used to verify our method, and the results were compared to an established estimation methodology. Our algorithm successfully estimated the respiratory effort, resistance, and elastance of the respiratory system, resulting in more robust performance and faster solver times than a previously proposed algorithm that used quadratic programming (QP) techniques. In a numerical simulation benchmark, the worst‐case errors for resistance and elastance were 25% and 23% for QP versus <0.1% and <0.1% for MIQP, whose solver times were 4.7 s and 0.5 s, respectively. This approach can estimate several breathing effort profiles and identify the respiratory system's mechanical properties in invasively ventilated critically ill patients. [ABSTRACT FROM AUTHOR]