Type 2 Diabetes Risk Prediction Using Deep Convolutional Neural Network Based-Bayesian Optimization

Diabetes mellitus is a long-term condition characterized by hyperglycemia. It could lead to plenty of difficulties. According to rising morbidity in recent years, the world’s diabetic patients will exceed 642 million by 2040, implying that one out of every ten persons will be diabetic. There is no doubt that this startling figure requires immediate attention from industry and academia to promote innovation and growth in diabetes risk prediction to save individuals’ lives. Due to its rapid development, deep learning (DL) was used to predict numerous diseases. However, DL methods still suffer from their limited prediction performance due to the hyperparameters selection and parameters optimization. Therefore, the selection of hyper-parameters is critical in improving classification performance. This study presents Convolutional Neural Network (CNN) that has achieved remarkable results in many medical domains where the Bayesian optimization algorithm (BOA) has been employed for hyperparameters selection and parameters optimization. Two issues have been investigated and solved during the experiment to enhance the results. The first is the dataset class imbalance, which is solved using Synthetic Minority Oversampling Technique (SMOTE) technique. The second issue is the model's poor performance, which has been solved using the Bayesian optimization algorithm. The findings indicate that the Bayesian based-CNN model superbases all the state-of-the-art models in the literature with an accuracy of 89.36%, F1-score of 0.88.6, and Matthews Correlation Coefficient (MCC) of 0.88.6.     

Kinetics of Thermal, Passive Oxidation of Nicalon Fibers

The oxidation of Nicalon fibers is a concern, because of its potential as a reinforcement of high-temperature composites, whose service conditions involve high-temperature, oxidizing environments. Two limiting types of oxidation mechanisms are often used to describe the kinetics: chemical-reaction-controlled oxidation, at small oxide thicknesses, and diffusion-controlled oxidation, at large oxide thicknesses. Neither mechanism can satisfactorily describe the intermediate region where the oxidation kinetics are controlled jointly by both the chemical reaction rate at the interface and the diffusion of oxygen through the oxide layer. To describe the entire oxidation process with a general relationship, one must consider all stages of the oxidation process, namely (i) adsorption of oxygen at the outer surface of the oxide, (ii) diffusion of oxygen from the outer surface toward the interface where oxidation occurs, and (iii) reaction at the interface to form a new layer of oxide. Previously, a very useful general relationship was derived for the oxidation kinetics for a flat plate, which could account for all three stages of oxidation. However, that equation is inadequate to describe the oxidation of cylindrical fibers, because the effective area for oxygen diffusion changes along the diffusion path and the oxidation interfacial area decreases as the oxide thickness increases for cylindrical fibers. In this paper, we have derived a general kinetic relationship for the oxidation of cylindrical fibers, which can account for all stages of oxidation. Comparison of the theory with experimental data of Nicalon fibers shows good agreement.     

Leveraging Transfer Learning for Spatio-Temporal Human Activity Recognition from Video Sequences

Human Activity Recognition (HAR) is an active research area due to its applications in pervasive computing, human-computer interaction, artificial intelligence, health care, and social sciences. Moreover, dynamic environments and anthropometric differences between individuals make it harder to recognize actions. This study focused on human activity in video sequences acquired with an RGB camera because of its vast range of real-world applications. It uses two-stream ConvNet to extract spatial and temporal information and proposes a fine-tuned deep neural network. Moreover, the transfer learning paradigm is adopted to extract varied and fixed frames while reusing object identification information. Six state-of-the-art pre-trained models are exploited to find the best model for spatial feature extraction. For temporal sequence, this study uses dense optical flow following the two-stream ConvNet and Bidirectional Long Short Term Memory (BiLSTM) to capture long-term dependencies. Two state-of-the-art datasets, UCF101 and HMDB51, are used for evaluation purposes. In addition, seven state-of-the-art optimizers are used to fine-tune the proposed network parameters. Furthermore, this study utilizes an ensemble mechanism to aggregate spatial-temporal features using a four-stream Convolutional Neural Network (CNN), where two streams use RGB data. In contrast, the other uses optical flow images. Finally, the proposed ensemble approach using max hard voting outperforms state-of-the-art methods with 96.30% and 90.07% accuracies on the UCF101 and HMDB51 datasets.     

Hydrodynamic drainage force in a highly confined geometry: role of surface roughness on different length scales

We measured the hydrodynamic drainage force of an aqueous, Newtonian liquid squeezed between two hydrophobic or two hydrophilic surfaces by means of the colloidal probe technique. We controlled the wettability, the roughness, the topology, and also the approaching velocity of the surfaces. We found that asperities on the surfaces caused an artificial decrease of the measured drainage force that must be considered by the interpretation of the force curves. Even considering the effect of asperities, our experimental results could be interpreted only with the aid of a partial slip model. Or else, interpreted assuming that the viscosity close to the surfaces is different from bulk. On patterned hydrophilic surfaces, we demonstrated that the drainage force depends not only on the overall surface roughness or micro structuring but also on the specific length scale of the surface nanostructures.     

Minimum Normal Force Principle Based Quantitive Optimizationof Clamping Forces for Thin Walled Part

Based on the stability criteria of workpiece-fixture system, quantitative optimization of clamping forces during precise machining process for thin walled part is studied considering the contact condition between wokpiece and locator, the contact mechanical model is achieved ,which is further been used to calculate the entire passive forces acting on the statically undetermined workpiece by means of the force screw theory as well as minimum norm force principle. Furthermore, a new methodology to optimize clamping forces is put forward, on the criteria of keeping the stability of workpiece during cutting process. By this way, the intensity of clamping forces is decreased dramatically, which will be most beneficial for improving the machining quality of thin-walled parts. Finally, a case study is used to support and validate the proposed model.     

Heterogeneous crystallization and composition dependence of optical parameters in Sn–Sb–Bi–Se chalcogenides

Bulk samples of Sn₁₀Sb_20−x Bi _x Se₇₀ (0 ≤ x ≤ 8) chalcogenide alloys were prepared by the conventional melt quenching technique. Thin films were prepared on well-cleaned glass substrates by thermal evaporation technique. X-ray diffraction studies revealed that the alloys with x = 0 and 2 at.% of Bi were amorphous, whereas the alloys with x = 4, 6, 8 at.% were crystalline. The crystalline phases are identified as due to the formation of Bi₂Se₃ and BiSe₂ phases. The microstructural and differential scanning calorimetric studies show the presence of these phases. A simple, straight forward procedure suggested by Swanepoel has been used to calculate the optical parameters, refractive index, and extinction coefficient. The optical gap for all the samples has been obtained from the Tauc plots. The variation in optical parameters for different Bi concentration has been explained on the basis of presence of defect states and the change in stoichiometry with the change in Bi concentration.     

Failure analysis of heat exchanger fluid vapor phase vent lines in a Polyester plant

A Polyester plant is using Proprietary heat exchange fluids (26.5% biphenyl, 73.5% diphenyloxide). The plant hot condensate tank, condensate feed tank, and heat exchanger vapor phase dead end and ventable lines (ASTM 106 Grade B) with isolation valves were suffering frequent leakages with severe gouging and pitting corrosion. Lustrous (shiny) coke deposit was prominent inside the cut sections due to coke formation by low boiler products. Some adhered brown iron oxide (rust) deposits were also visible. A sludge sample was analyzed by XRD which revealed that carbon was the major constituent with γ-Maghemite-C (an iron oxide) as second mineral in order of abundance. The failure was diagnosed as severe localized corrosion and gouging at temperatures of about 200 °C and 53 psi by non-condensable low boilers decomposed corrosive products. The remedy was suggested which has been validated over the 2 years trouble free performance.     

Structural,Mechanical and Dielectric Properties of Microwave-Assisted High-Energy Ball Milling Synthesis of Hydroxyapatite

Abstract

In the present work, the effect of high-energy ball milling (HEBM) of starting precursors of hydroxyapatite (Ca₁₀(PO₄)₆(OH)₂)/HA system on the processing temperature, morphological, mechanical, and electrical properties are highlighted. Calcination and sintering of HA system were carried out in microwave furnace. XRD study confirmed the evolution of HA phase and EDX analysis confirmed Ca/P ratio ～1.65. Grain size of HA samples, synthesized by using microwave-assisted HEBM technique was found to be in the order of 54–75?nm. Enhanced dielectric and mechanical properties were obtained in HA samples.     

Kinetics of Thermal Dehydroxylation and Carbonation of Magnesium Hydroxide

The kinetics of simultaneous dehydroxylation and carbonation of precipitated Mg(OH)₂ were studied using isothermal and nonisothermal thermogravimetric analyses. Specimens were analyzed using X-ray diffraction, transmission electron microscopy, and through measurements of the volume of carbon dioxide evolved in a subsequent reaction with hydrochloric acid. From 275° to 475°C, the kinetics of isothermal dehydroxylation in helium were best fit to a contracting-sphere model, yielding an activation energy of 146 kJ/mol, which was greater than values reported in the literature for isothermal dehydroxylation under vacuum (53–126 kJ/mol). The carbonation kinetics were complicated by the fact that dehydroxylation occurred simultaneously. The overall kinetics also could be fit to a contracting-sphere model, yielding a net activation energy of 304 kJ/mol. The most rapid carbonation kinetics occurred near 375°C. At this temperature, Mg(OH)₂ underwent rapid dehydroxylation and subsequent phase transformation, whereas thermodynamics favored the formation of carbonate. During carbonation, MgCO₃ precipitated on the surface of disrupted Mg(OH)₂ crystals acting as a kinetic barrier to both the outward diffusion of H₂O and the inward diffusion of CO₂.