Compatible natural gradient policy search

Machine Learning - Trust-region methods have yielded state-of-the-art results in policy search. A common approach is to use KL-divergence to bound the region of trust resulting in a natural...     

Diagnosis of breast cancer based on modern mammography using hybrid transfer learning

Breast cancer is a common cancer in women. Early detection of breast cancer in particular and cancer, in general, can considerably increase the survival rate of women, and it can be much more effective. This paper mainly focuses on the transfer learning process to detect breast cancer. Modified VGG (MVGG) is proposed and implemented on datasets of 2D and 3D images of mammograms. Experimental results showed that the proposed hybrid transfer learning model (a fusion of MVGG and ImageNet) provides an accuracy of 94.3%. On the other hand, only the proposed MVGG architecture provides an accuracy of 89.8%. So, it is precisely stated that the proposed hybrid pre-trained network outperforms other compared Convolutional Neural Networks. The proposed architecture can be considered as an effective tool for radiologists to decrease the false negative and false positive rates. Therefore, the efficiency of mammography analysis will be improved.

     

Fabrication of asymmetric zinc oxide/carbon nanotubes coated polysulfone photocatalytic nanocomposite membrane for fouling mitigation

Membrane filtration is a favorable option in water reclamation from contaminated water source, nevertheless, inevitable membrane fouling which greatly shortens membrane longevity and separation efficiency. The paper aimed to mitigate the membrane fouling through the formation of an asymmetric PSF-ZnO/CNTs photocatalytic nanocomposite membrane. Instead of direct blending the photocatalyst into polymer matrix, the asymmetric nanocomposite membrane was prepared with prior formation of a self-assembled ZnO/CNTs photocatalyst layer through wet-processing technique followed with coating of PSF support layer via phase inversion method. The morphology of the nanocomposite membrane was characterized to confirm the formation of the asymmetric structure. The effect of ZnO/CNTs photocatalyst loading on the pore characteristic and antifouling properties of the PSF-ZnO/CNTs nanocomposite membrane in dye remediation were assessed. The incorporation of ZnO/CNTs layer was found to endows the membrane with ability to photodegrade methylene blue. The PSF-ZnO/CNTs membrane with 0.038 g ZnO/CNTs photocatalyst loading (M5) showed the greatest flux recovery ratio (98.46%) and the lowest irreversible fouling ratio (1.54%) while exhibited decent water permeability about 29.99 L/m²h without compromise the methylene blue rejection rate (91.04%).     

Recent large-diameter tunnel construction in Singapore using sprayed concrete linnings

Sprayed Concrete Lining (SCL) or the New Austrian Tunnelling Method (NATM) has had limited exposure in Singapore. Previously this has been restricted to excavations of less than 6 metres inside diameter in the boulder clay and decomposed granite. For the North East Line Project (NELP) a section of SCL tunnel was constructed as part of the tunnelling works on Contract 710. This paper describes the design, planning, and construction issues of twin 8.4-metre inside diameter crossover tunnels spaced less than one diameter apart constructed in the Jurong Formation. In addition, the construction of a SCL tunnel in the Old Alluvium, is also described. This tunnel is an advance tunnel for the future Deep Tunnel Sewage Scheme (DTSS) and is being constructed as part of the NELP Contract 705 Works. Details of the Jurong Formation and Old Alluvium are given together with their geotechnical characteristics, which played a major role in the selection of temporary support methods used and the behaviour of the ground during tunnelling. The planning and design issues concentrate on aspects to ensure overall integrity and stability of both SCL projects. The success of various construction techniques utilised to construct the tunnels safely is discussed and a summary of the ground movements is presented.     

Adaptation to climate change for food security in the lower Mekong Basin

Variability in water cycles driven by climate change is considered likely to impact rice production in the near future. Rice is the main staple food for the population in the lower Mekong Basin and the demand for food is expected to grow due to increase in population. This paper examines the impact of climate change on rice production in the lower Mekong Basin, evaluates some widely used adaptation options, and analyses their implications for overall food security by 2050. Climate change data used in the study are the future climate projection for two IPCC SRES scenarios, A2 and B2, based on ECHAM4 General Circulation Model downscaled to the Mekong region using the PRECIS (Providing Regional Climates for Impact Studies) system. In general, the results suggest that yield of rainfed rice may increase significantly in the upper part of the basin in Laos and Thailand and may decrease in the lower part of the basin in Cambodia and Vietnam. Irrigated rice may not be affected by climate change if increased irrigation requirements are met. Negative impact on the yield of rainfed rice can be offset and net increase in yield can be achieved by applying widely used adaptation options such as changing planting date, supplementary irrigation and increased fertilizer input. Analysis of the projected production, considering population growth by 2050, suggests that food security of the basin is unlikely to be threatened by the increased population and climate change, excluding extreme events such as sea level rise and cyclones.     

Multi-extremum-modified response basis model for nonlinear response prediction of dynamic turbine blisk

For the nonlinear dynamic analyses of complex mechanical components, it is necessary to apply efficient modeling framework to reduce computational burden. The accurate surrogate model for approximating the nonlinear responses of several failures is a vital issue to provide robust and safe design conditions in complex engineering applications. In this paper, two different Modified multi-extremum Response Surface basis Models (MRSM) are proposed for dynamic nonlinear responses of failure capacities for turbine blisk responses. The proposed MRSM is established using two regression processes including regressed the input variables by linear or exponential basis functions in first calibrating phase and regressed the second-order polynomial basis function using inputs data provided by first stage in second calibrating procedure. A sensitivity analysis using MRSM is proposed to consider the variation of input variables on the nonlinear responses. In the sensitivity analysis procedure, the effects of input variables are evaluated using the calibrating results given from the first regressed process. To evaluate the performance of the proposed MRSM, three multi-extremum failure modes including radial deformation of compressor blisk, maximum strain, and stress of compressor blade and disk are considered. the prediction of MRSM of nonlinear responses for Thermal-fluid–structure system with dynamical nonlinear finite-element analyses is compared with response surface method (RSM) and artificial neural network (ANN). The predicted results of modeling approaches showed that the sensitivity analysis based on MRSM accurately provided the effective degree for input variables. The gas temperature has the highest effects on nonlinear responses of turbine blisk which is followed by angular speed and material density. The MRSM combined with basic exponential function performs better than other models, while the MRSM coupled with linear function is more accurate than ANN and RSM. The proposed MRSM models have illustrated the accurate and efficient framework for approximating dynamic structural analysis of complex components.

     

Upconversion luminescence of Gd2O3:Er3+ and Gd2O3:Er3+/silica nanophosphors fabricated by EDTA combustion method

Gd_2O_3:Er~(3+) nanophosphors were fabricated by the combustion method in presence of Na_2 ethylene diamine tetra acetic acid(EDTA-Na_2) as fuel at not high temperature(≤350℃) within a very short time of 5 min.The added concentration of Er~(3+)ions in Gd_2O_3 matrix was changed from 0.5 mol% to 5.0 mol%.The X-ray diffraction pattern of samples indicates the monoclinic structure of Gd_2O_3:Er3+.The morphology and chemical composition analysis of the Gd_2O_3:Er~(3+) samples are characterized by a field emission scanning electron microscope(FESEM) and a Fourier-transform infrared spectrometer(FTIR).The photoluminescence(PL),photo luminescence excitation(PLE) and upconversion(UC) at room temperature of the prepared materials with different concentrations of Er~(3+) were investigated.The PL of Gd_2O_3:Er~(3+)nanomaterials are shown in visible at 545,594,623,648,688 nm under excitation at 275 nm.The emission bands from transitions of Er~(3+) from ~2P_(3/2) to ~4F_(9/2) are observed,UC luminescent spectra of the Gd_2O_3:Er~(3+)/silica nanocomposites under 976 nm excitation show the bands at 548 and 670 nm.The influence of excitation power at 980 nm for transitions were measured and calculated.The results indicate that the upconversion process of Gd_2O_3:Er~(3+)/silica is two photons absorption mechanism.The low temperature dependence of UC luminescent intensities of the main bands of Gd_2O_3:Er~(3+)was investigated towards development of a nanotemperature sensor in the range of 10-300 K.     

Sulfonamide functionalized amino acid-based pH- and temperature-sensitive biodegradable injectable hydrogels: Synthesis,physicochemical characterization and in vivo degradation kinetics

In this research, a novel pH and temperature-sensitive biodegradable oligomer serin-b-poly(lactide)-b-poly (ethylene glycol)-b-poly(lactide)-b- oligomer serin (OS-PLA–PEG–PLA-OS) pentablock copolymer was synthesized with modification of serine to sulfonamide group. The properties of the different pentablock copolymer structures were detail characterized. The copolymer solution easily flowed at high pH and changed to gel state at physiological conditions (37°C, pH 7.4). The sol to gel behavior of these block copolymer solutions were controllable by tuning the length of PEG segment, PLA/PEG block ratio, the molecular weight and the concentration of the pentablock copolymer. The in vivo gelation of the copolymer solution was investigated. The stable gel was formed after injection into the back of the adult male Mus musculus Swiss Albino mice. The degradation process was observed after 6 weeks and showed that the hydrogel depot was bioresorbable without any detection of inflammation sign. The results in this study suggest that the OS-PLA–PEG–PLA-OS pentablock copolymer has the potential application as an injectable delivery vehicle for therapeutic drugs that easily denatured at low-pH condition.     

“Don’t Neglect the User!” – Identifying Types of Human-Chatbot Interactions and their Associated Characteristics

Information Systems Frontiers - Interactions with conversational agents (CAs) become increasingly common in our daily life. While research on human-CA interactions provides insights into the role...