Compact Deep Color Features for Remote Sensing Scene Classification

Neural Processing Letters - Aerial scene classification is a challenging problem in understanding high-resolution remote sensing images. Most recent aerial scene classification approaches are based...     

Ionic liquids for CO2 electrochemical reduction

Electrochemical reduction of CO₂ is a novel research field towards a CO₂-neutral global economy and combating fast accelerating and disastrous climate changes while finding new solutions to store renewable energy in value-added chemicals and fuels. Ionic liquids (ILs), as medium and catalysts (or supporting part of catalysts) have been given wide attention in the electrochemical CO₂ reduction reaction (CO₂RR) due to their unique advantages in lowering overpotential and improving the product selectivity, as well as their designable and tunable properties. In this review, we have summarized the recent progress of CO₂ electro-reduction in IL-based electrolytes to produce higher-value chemicals. We then have highlighted the unique enhancing effect of ILs on CO₂RR as templates, precursors, and surface functional moieties of electrocatalytic materials. Finally, computational chemistry tools utilized to understand how the ILs facilitate the CO₂RR or to propose the reaction mechanisms, generated intermediates and products have been discussed.     

Cross-Sectional Associations of Food Consumption with Plasma Fatty Acid Composition and Estimated Desaturase Activities in Finnish Children

Plasma fatty acid (FA) composition is known to be an indicator of dietary fat quality, but the associations of other dietary factors with plasma FA composition remain unknown in children. We investigated the cross-sectional associations of food consumption with the proportions of FA and estimated desaturase activities in plasma cholesteryl esters (CE) and phospholipids (PL) among children. The subjects were a population sample of 423 children aged 6–8 years examined at baseline of The Physical Activity and Nutrition in Children (PANIC) Study. We assessed food consumption by food records and plasma FA composition by gas chromatography. We used linear regression models adjusted for age, sex, physical activity and total energy intake to analyze the associations. A higher consumption of vegetable oil-based margarine (fat 60–80 %) was associated with a higher proportion of linoleic and α-linolenic acids in plasma CE and PL. A higher consumption of high-fiber grain products was related to a lower proportion of oleic acid in CE and PL. The consumption of candy was directly associated with the proportion of palmitoleic and oleic acid in plasma CE. The consumption of vegetable oil-based margarine was inversely associated with estimated stearoyl-CoA-desaturase activity in plasma CE and PL and the consumption of candy was directly related to it in plasma CE. The results of our study suggest that plasma FA composition is not only a biomarker for dietary fat quality but also reflects the consumption of high-fiber grain products and foods high in sugar among children.     

Neural and statistical classifiers-taxonomy and two case studies

Pattern classification using neural networks and statistical methods is discussed. We give a tutorial overview in which popular classifiers are grouped into distinct categories according to their underlying mathematical principles; also, we assess what makes a classifier neural. The overview is complemented by two case studies using handwritten digit and phoneme data that test the performance of a number of most typical neural-network and statistical classifiers. Four methods of our own are included: reduced kernel discriminant analysis, the learning k-nearest neighbors classifier, the averaged learning subspace method, and a version of kernel discriminant analysis.     

Visualization of solvation structures in liquid mixtures

Spatial distribution functions of atomic densities. SDFs, have been proposed as a natural starting point for analysis of local molecular structure in liquids and solutions. The local structure in these systems is often complex and this is reflected in the fact that SDFs can be difficult to visualize. Among the different methods that can be used to visualize SDFs we discuss 3D isodensity surfaces, cross-sections, and 'comic book' animations. We also discuss the possibility of a simultaneous visualization of SDFs and other 3D fields, such as the electron density. These techniques are all intended to emphasize and bring out aspects of SDFs that promote a further understanding of the local molecular structure. OpenGL-based software has been used under X-Windows to implement these techniques, and we argue that high-quality molecular graphics need not be expensive. Data from a molecular dynamics simulation of an equimolar binary mixture of water and acetonitrile have been used to illustrate the discussion.     

A numerical Hartree-Fock program for diatomic molecules

This paper describes an algorithm and a computer program which solves numerically (virtually exactly) equations of the restricted open-shell Hartree-Fock and Hartree-Fock-Slater model for diatomic molecules     

Scale coding bag of deep features for human attribute and action recognition

Most approaches to human attribute and action recognition in still images are based on image representation in which multi-scale local features are pooled across scale into a single, scale-invariant encoding. Both in bag-of-words and the recently popular representations based on convolutional neural networks, local features are computed at multiple scales. However, these multi-scale convolutional features are pooled into a single scale-invariant representation. We argue that entirely scale-invariant image representations are sub-optimal and investigate approaches to scale coding within a bag of deep features framework. Our approach encodes multi-scale information explicitly during the image encoding stage. We propose two strategies to encode multi-scale information explicitly in the final image representation. We validate our two scale coding techniques on five datasets: Willow, PASCAL VOC 2010, PASCAL VOC 2012, Stanford-40 and Human Attributes (HAT-27). On all datasets, the proposed scale coding approaches outperform both the scale-invariant method and the standard deep features of the same network. Further, combining our scale coding approaches with standard deep features leads to consistent improvement over the state of the art.     

Angiogenic Effects and Crosstalk of Adipose-Derived Mesenchymal Stem/Stromal Cells and Their Extracellular Vesicles with Endothelial Cells

Adipose-derived mesenchymal stem/stromal cells (ASCs) are an adult stem cell population able to self-renew and differentiate into numerous cell lineages. ASCs provide a promising future for therapeutic angiogenesis due to their ability to promote blood vessel formation. Specifically, their ability to differentiate into endothelial cells (ECs) and pericyte-like cells and to secrete angiogenesis-promoting growth factors and extracellular vesicles (EVs) makes them an ideal option in cell therapy and in regenerative medicine in conditions including tissue ischemia. In recent angiogenesis research, ASCs have often been co-cultured with an endothelial cell (EC) type in order to form mature vessel-like networks in specific culture conditions. In this review, we introduce co-culture systems and co-transplantation studies between ASCs and ECs. In co-cultures, the cells communicate via direct cell–cell contact or via paracrine signaling. Most often, ASCs are found in the perivascular niche lining the vessels, where they stabilize the vascular structures and express common pericyte surface proteins. In co-cultures, ASCs modulate endothelial cells and induce angiogenesis by promoting tube formation, partly via secretion of EVs. In vivo co-transplantation of ASCs and ECs showed improved formation of functional vessels over a single cell type transplantation. Adipose tissue as a cell source for both mesenchymal stem cells and ECs for co-transplantation serves as a prominent option for therapeutic angiogenesis and blood perfusion in vivo.     

Electrochemical reduction of oxygen on thin-film Pt electrodes in acid solutions

The electrochemical reduction of oxygen on thin-film platinum electrodes in 0.1 M HClO₄ and 0.05 M H₂SO₄ solutions has been investigated using the rotating disk electrode (RDE) method. Thin films of Pt (0.25-20 nm thick) were prepared by vacuum evaporation onto glassy carbon substrate. The surface morphology of Pt films was examined by transmission electron microscopy (TEM). The specific activity of O₂ reduction was higher in HClO₄ and decreased with decreasing film thickness. In H₂SO₄, the specific activity was lower and appeared to be independent of the Pt loading. The values of Tafel slopes close to −120 mV dec⁻¹ in high current density range and −60 mV dec⁻¹ in low current density range were obtained for all electrodes in both solutions, indicating that the mechanism of O₂ reduction is the same for thin-film electrodes as for bulk Pt. The number of electrons transferred per O₂ molecule was close to four for all thin Pt films studied.