Accurate ahead evapotranspiration forecasting is crucial for irrigation planning, for wetlands, agricultural and forest habitats preservation, and for water resource management. Deep learning ...algorithms can be used to develop effective forecasting models of ahead evapotranspiration. In this study, three Recurrent Neural Network-based models were built for the prediction of short term ahead actual evapotranspiration. Two variants of each model were developed changing the employed algorithm, selecting between long short-term memory (LSTM) and nonlinear autoregressive network with exogenous inputs (NARX), while the modeling was performed in the context of an ensemble approach. The prediction models were trained and tested using data from two sites with different climates: Cypress Swamp, southern Florida, and Kobeh Valley, central Nevada. With reference to the subtropical climatic conditions of South Florida, LSTM models proved to be more accurate than NARX models, while some exogenous variables such as sensible heat flux and relative humidity did not affect the results significantly. An increase of the forecast horizon from 1 to 7 days resulted in a slight reduction in the accuracy of both the LSTM- and NARX based models. Considering instead the semi-arid climate of Central Nevada, NARX models generally provided more accurate results, which were only slightly affected by relative humidity, sensible heat flux, and forecast horizon. On the other hand, LSTM models performance decayed if sensible heat flux and relative humidity were neglected, and if the forecast horizon was increased from 1 to 7 days. Deep learning-based models can provide very accurate predictions of actual evapotranspiration, but the performance of the models can be significantly affected by local climatic conditions.
Display omitted
•Deep learning-based models can give accurate forecasting of actual evapotranspiration.•Models were developed using an ensemble approach.•Forecasting models were compared in subtropical and semiarid climatic conditions.•The performance of the models is depending on the algorithm, forecast horizon, and climate.
Phthalates and bisphenol A (BPA) are ubiquitous contaminants identified as endocrine disruptors. Phthalates are worldwide used as plasticizers, in particular to improve the mechanical properties of ...polymers such as polyvinyl chloride. Because they are not chemically bound to the polymer, they tend to leach out with time and use. Di-2-ethylhexyl phthalate (DEHP) and di-n-butyl phthalate (DnBP) are the two most common phthalates. BPA is an estrogenic compound used to manufacture polycarbonate containers for food and drink, including baby bottles. It can migrate from container into foods, especially at elevated temperatures. Diet is a predominant source of exposure for phthalates and BPA, especially for infants. The aim of this study was to test the presence of DEHP, DnBP, and BPA in infant formulas. DEHP, DnBP, and BPA concentrations were measured in 22 liquid and 28 powder milks by gas chromatography with flame ionization detection and high performance liquid chromatography with fluorimetric detection, respectively. DEHP concentrations in our samples were between 0.005 and 5.088 μg/g (median 0.906 μg/g), DnBP concentrations were between 0.008 and 1.297 μg/g (median 0.053 μg/g), and BPA concentrations were between 0.003 and 0.375 μg/g (median 0.015 μg/g). Concentrations of the investigated contaminants in liquid and powder milks were not significantly different, even though samples were packed in different types of containers. These data point out potential hazards for infants fed with baby formulas. Contamination seems more related to the production of formulas than to a release from containers.
Selection of patients for preoperative treatment in rectal cancer is controversial. The new 2020 National Institute for Health and Care Excellence (NICE) guidelines, consistent with the National ...Comprehensive Cancer Network guidelines, recommend preoperative radiotherapy for all patients except for those with radiologically staged T1–T2, N0 tumours. We aimed to assess outcomes in non-irradiated patients with rectal cancer and to stratify results on the basis of NICE criteria, compared with known MRI prognostic factors now omitted by NICE.
For this retrospective cohort study, we identified patients undergoing primary resectional surgery for rectal cancer, without preoperative radiotherapy, at Basingstoke Hospital (Basingstoke, UK) between Jan 1, 2011, and Dec 31, 2016, and at St Marks Hospital (London, UK) between Jan 1, 2007, and Dec 31, 2017. Patients with MRI-detected extramural venous invasion, MRI-detected tumour deposits, and MRI-detected circumferential resection margin involvement were categorised as MRI high-risk for recurrence (local or distant), and their outcomes (disease-free survival, overall survival, and recurrence) were compared with patients defined as high-risk according to NICE criteria (MRI-detected T3+ or MRI-detected N+ status). Kaplan-Meier and Cox proportional hazards analyses were used to compare the groups.
378 patients were evaluated, with a median of 66 months (IQR 44–95) of follow up. 22 (6%) of 378 patients had local recurrence and 68 (18%) of 378 patients had distant recurrence. 248 (66%) of 378 were classified as high-risk according to NICE criteria, compared with 121 (32%) of 378 according to MRI criteria. On Kaplan-Meier analysis, NICE high-risk patients had poorer 5-year disease-free survival compared with NICE low-risk patients (76% 95% CI 70–81 vs 87% 80–92; hazard ratio HR 1·91 95% CI 1·20–3·03; p=0·0051) but not 5-year overall survival (80% 74–84 vs 88% 81–92; 1·55 0·94–2·53; p=0·077). MRI criteria separated patients into high-risk versus low-risk groups that predicted 5-year disease-free survival (66% 95% CI 57–74 vs 88% 83–91; HR 3·01 95% CI 2·02–4·47; p<0·0001) and 5-year overall survival (71% 62–78 vs 89% 84–92; 2·59 1·62–3·88; p<0·0001). On multivariable analysis, NICE risk assessment was not associated with either disease-free survival or overall survival, whereas MRI criteria predicted disease-free survival (HR 2·74 95% CI 1·80–4·17; p<0·0001) and overall survival (HR 2·44 95% CI 1·51–3·95; p=0·00027). 139 NICE high-risk patients who were defined as low-risk based on MRI criteria had similar disease-free survival as 118 NICE low-risk patients; therefore, 37% (139 of 378) of patients in this study cohort would have been overtreated with NICE 2020 guidelines. Of the 130 patients defined as low-risk by NICE guidelines, 12 were defined as high-risk on MRI risk stratification and would have potentially been missed for treatment.
Compared to previous guidelines, implementation of the 2020 NICE guidelines will result in significantly more patients receiving preoperative radiotherapy. High-quality MRI selects patients with good outcomes (particularly low local recurrence) without radiotherapy, with little margin for improvement. Overuse of radiotherapy could occur with this unselective approach. The high-risk group, with the most chance of benefiting from preoperative radiotherapy, is not well selected on the basis of NICE 2020 criteria and is better identified with proven MRI prognostic factors (extramural venous invasion, tumour deposits, and circumferential resection margin).
None.
In recent years, the growing impact of climate change on surface water bodies has made the analysis and forecasting of streamflow rates essential for proper planning and management of water ...resources. This study proposes a novel ensemble (or hybrid) model, based on the combination of a Deep Learning algorithm, the Nonlinear AutoRegressive network with eXogenous inputs, and two Machine Learning algorithms, Multilayer Perceptron and Random Forest, for the short-term streamflow forecasting, considering precipitation as the only exogenous input and a forecast horizon up to 7 days. A large regional study was performed, considering 18 watercourses throughout the United Kingdom, characterized by different catchment areas and flow regimes. In particular, the predictions obtained with the ensemble Machine Learning-Deep Learning model were compared with the ones achieved with simpler models based on an ensemble of both Machine Learning algorithms and on the only Deep Learning algorithm. The hybrid Machine Learning-Deep Learning model outperformed the simpler models, with values of R
above 0.9 for several watercourses, with the greatest discrepancies for small basins, where high and non-uniform rainfall throughout the year makes the streamflow rate forecasting a challenging task. Furthermore, the hybrid Machine Learning-Deep Learning model has been shown to be less affected by reductions in performance as the forecasting horizon increases compared to the simpler models, leading to reliable predictions even for 7-day forecasts.
Foot ulcers account for 15% of comorbidities associated with diabetes. Presently, no device allows the status of foot ulcers to be continuously monitored when patients are not hospitalized. In this ...study, we describe a temperature and a pH sensor capable of monitoring diabetic foot and venous leg ulcers developed in the frame of the seventh framework program European Union project SWAN-iCare (smart wearable and autonomous negative pressure device for wound monitoring and therapy). Temperature is measured by exploiting the variations in the electrical resistance of a nanocomposite consisting of multiwalled carbon nanotubes and poly(styrene-b-(ethylene-
-butylene)-b-styrene). The pH sensor used a graphene oxide (GO) layer that changes its electrical potential when pH changes. The temperature sensor has a sensitivity of ~85 Ω/°C in the range 25°C-50°C and a high repeatability (maximum standard deviation of 0.1% over seven repeated measurements). For a GO concentration of 4 mg/mL, the pH sensor has a sensitivity of ~42 mV/pH and high linearity (R2=0.99).
A patient at the fifth stage of chronic kidney disease usually needs dialysis or transplantation. The adequacy of dialysis can be assessed by the dimensionless parameter Kt/V , which depends on the ...concentration of urea in the blood. Kt/V is usually measured before and after treatment, or by an indirect measurement of dialysis fluid conductivity. For the real-time monitoring of the urea concentration during dialysis, this paper proposes a disposable biosensor that combines a pH sensitive reduced graphene oxide layer functionalized with 4-amino benzoic acid and urease. Urease is immobilized on reduced graphene oxide and catalyzes the hydrolysis of urea into carbon dioxide and ammonia. This catalysis increases the pH value locally and elicits a sensor response. Urea concentration was assessed in the plasma of five dialyzed patients using a potentiometric measurement. The response time was 120 s and the error was 6 ± 3% compared with standard clinical laboratory analysis.
AbstractA two-phase air-water shear flow generated by a submerged jet and characterized by a relevant void fraction and laden with fluorescent particles has been investigated experimentally by means ...of particle image velocimetry (PIV) combined with the laser-induced fluorescence (LIF). Two air-water data sets have been recorded in order to explore two different regimes, namely two-way coupling and four-way coupling, according to air–water interactions. A novel robust discrimination algorithm based on the geometric features and pixel intensity of tracers and bubbles has been introduced and validated. A statistical analysis of the flow field for the two phases and of the bubble-phase geometric features and distribution has been conducted. The increase of the air-flow rate involves an upward orientation of the jet due to the interaction between the air bubbles’ buoyancy forces and the water, with a turbulence increase. Furthermore, greater interactions between air bubbles grouped in clusters occur, backed up by less variation of the air bubbles’ shape and orientation passing from the jet zone to the zone above the same. The results of this study have led to better knowledge of bubble–water and bubble–bubble interactions and to an estimation of the effects of air bubbles in the shear layer of a bubbly water jet.
Heart failure (HF) is the main cause of mortality worldwide, particularly in the elderly. N-terminal pro-brain natriuretic peptide (NT-proBNP) is the gold standard biomarker for HF diagnosis and ...therapy monitoring. It is determined in blood samples by the immunochemical methods generally adopted by most laboratories. Saliva analysis is a powerful tool for clinical applications, mainly due to its non-invasive and less risky sampling. This study describes a validated analytical procedure for NT-proBNP determination in saliva samples using a commercial Enzyme-Linked Immuno-Sorbent Assay. Linearity, matrix effect, sensitivity, recovery and assay-precision were evaluated. The analytical approach showed a linear behaviour of the signal throughout the concentrations tested, with a minimum detectable dose of 1 pg/mL, a satisfactory NT-proBNP recovery (95-110%), and acceptable precision (coefficient of variation ≤ 10%). Short-term (3 weeks) and long-term (5 months) stability of NT-proBNP in saliva samples under the storage conditions most frequently used in clinical laboratories (4, - 20, and - 80 °C) was also investigated and showed that the optimal storage conditions were at - 20 °C for up to 2.5 months. Finally, the method was tested for the determination of NT-proBNP in saliva samples collected from ten hospitalized acute HF patients. Preliminary results indicate a decrease in NT-proBNP in saliva from admission to discharge, thus suggesting that this procedure is an effective saliva-based point-of-care device for HF monitoring.
The deformation of air bubbles in a liquid flow field is of relevant interest in phenomena such as cavitation, air entrainment, and foaming. In complex situations, this problem cannot be addressed ...theoretically, while the accuracy of an approach based on Computational Fluid Dynamics (CFD) is often unsatisfactory. In this study, a novel approach to the problem is proposed, based on the combined use of a shadowgraph technique, to obtain experimental data, and some machine learning algorithms to build prediction models. Three models were developed to predict the equivalent diameter and aspect ratio of air bubbles moving near a plunging jet. The models were different in terms of their input variables. Five variants of each model were built, changing the implemented machine learning algorithm: Additive Regression of Decision Stump, Bagging, K-Star, Random Forest and Support Vector Regression. In relation to the prediction of the equivalent diameter, two models provided satisfactory predictions, assessed on the basis of four different evaluation metrics. The third model was slightly less accurate in all its variants. Regarding the forecast of the bubble’s aspect ratio, the difference in the input variables of the prediction models shows a greater influence on the accuracy of the results. However, the proposed approach proves to be promising to address complex problems in the study of multi-phase flows.
The disturbance of protein O-GlcNAcylation is emerging as a possible link between altered brain metabolism and the progression of neurodegeneration. As observed in brains with Alzheimer's disease ...(AD), flaws of the cerebral glucose uptake translate into reduced protein O-GlcNAcylation, which promote the formation of pathological hallmarks. A high-fat diet (HFD) is known to foster metabolic dysregulation and insulin resistance in the brain and such effects have been associated with the reduction of cognitive performances. Remarkably, a significant role in HFD-related cognitive decline might be played by aberrant protein O-GlcNAcylation by triggering the development of AD signature and mitochondrial impairment. Our data support the impairment of total protein O-GlcNAcylation profile both in the brain of mice subjected to a 6-week high-fat-diet (HFD) and in our in vitro transposition on SH-SY5Y cells. The reduction of protein O-GlcNAcylation was associated with the development of insulin resistance, induced by overfeeding (i.e., defective insulin signaling and reduced mitochondrial activity), which promoted the dysregulation of the hexosamine biosynthetic pathway (HBP) flux, through the AMPK-driven reduction of GFAT1 activation. Further, we observed that a HFD induced the selective impairment of O-GlcNAcylated-tau and of O-GlcNAcylated-Complex I subunit NDUFB8, thus resulting in tau toxicity and reduced respiratory chain functionality respectively, highlighting the involvement of this posttranslational modification in the neurodegenerative process.