Search

Proanthocyanidin liposome modified by polyethylene glycol (PEG) was prepared by reverse-phase evaporation method. To obtain the higher encapsulation efficiency of proanthocyanidin liposomes, several factors, including the mass ratio of lecithin to cholesterol and the addition of surfactants and PEG, were investigated. The liposomes then can be evaluated by encapsulation efficiency, the distribution of particle size and storage stability. The results showed that the encapsulation efficiency of the liposomes was around 77.28 %, and the average particle size of the liposomes was 144.9 ± 0.1 nm. In addition, proanthocyanidin could be controlled to release on simulated skin in vitro. These data confirmed that the liposomes could be employed to extend proanthocyanidin's shelf life and to provide more convenience and better performance for its usage in cosmetics. [ABSTRACT FROM AUTHOR]

Harmony and preference are two important factors in constructing color combination for the purpose of color design. We presented a method for generating harmonious color scheme in a previous project. As an extension to the project, we propose here a method for generating color schemes that are based on user's preference. We considered color combinations as a set of color relations rather than a summation of specific color components (color elements); and we derived a model of six-variable color relations, representative value of hue, representative value of saturation, representative value of lightness, span of hue, span of saturation, and span of lightness. With this model, color combinations can be expressed as specific parameters of the six-variable model. Individual preferences are represented as certain parameters, from which new color schemes can be reconstructed to meet individual requirements. An interactive system for generating preferential color schemes is built with all these principles integrated. As a color design toolkit for user study, the system accesses user's preference features for color combinations, and provides new color schemes that fit the user's need. User study shows that both professionals and novices can generally acquire preferential color schemes of their own. © 2013 Wiley Periodicals, Inc. Col Res Appl, 40, 147-156, 2015 [ABSTRACT FROM AUTHOR]

Different kinds of organic montmorillonite (OMMT) were added into Polyamide 1010 (PA1010)/nanometer calcium carbonate (nano-CaCO) nanocomposites to prepare PA1010 ternary nanocomposites by melt blending method via twin screw extruder, which were characterized by mechanics performance testing, scanning electron microscope (SEM), wide X-ray diffraction (WAXD) and differential scanning calorimetry meter (DSC). The results revealed OMMT of larger interlayer spacing weaken strong adsorption force between nano-CaCO and matrix so as to boost tensile strength and flexural strength. OMMT of small interlayer spacing could produce more yields and empties in order to make matrix appear plastic fracture characteristics. By SEM images cross-section showed matrix toughness had been improved. Meanwhile more forward translation retrained the formation of α-form while γ-form was facilitated through WAXD, which toughened matrix consequently. Smaller layer spacing OMMT at the same content is obviously able to decrease crystallinity and T at little OMMT content, which exhibited superior dispersion of nanofillers. [ABSTRACT FROM AUTHOR]

We present an interactive visualization tool for generating color schemes that can be used by both professional artists and novice designers to improve their design efficiency. The method is developed on two color harmony principles of familial factors and rhythmic spans from conventional color theories. The interactive tool makes use of these principles for users to interactively generate colors palettes containing harmonious combinations in the hue, saturation, and lightness color space in as few steps as possible. A pilot study with 20 participants has shown that this tool provides efficient workflow, color-scheme harmony, and intuitive interaction, demonstrating the usefulness of the method. This work contributes to an artistic-theory driven harmonious color-scheme generation and to interactive user interfaces visualizing color relationships. © 2012 Wiley Periodicals, Inc. Col Res Appl, 39, 70-78, 2014 [ABSTRACT FROM AUTHOR]

Self-attention can capture long-distance dependencies and is widely used in semantic segmentation. Existing methods mainly use two kinds of self-attentions, i.e., spatial attention and channel attention, which can capture the relations in H W dimension (image plane, height and width) and C dimension (channels), respectively. Very little research investigates self-attention along other dimensions, which can potentially improve the segmentation performance. In this work, we investigate the self-attentions along all the possible dimensions { H , W , C , H W , H C , C W , H W C }. Then we explore the aggregation of all the possible self-attentions. We apply the neural architecture search (NAS) technique to achieve optimal aggregation. Specifically, we carefully design (1) the search space and (2) the optimization method. For (1), we introduce a building block, a basic self-attention search unit (BSU), which can model self-attentions along all the dimensions. And the search space contains within-BSU and cross-BSU operations. In addition, we propose an attention-map splitting method, which can reduce the computations by 1/3. For (2), we apply an efficient differentiable optimization method to search the optimal aggregation. We conduct extensive experiments on Cityscapes and ADE20K datasets. The results show the effectiveness of the proposed method, and we achieve very competitive performance against state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Polyamide 11 (PA 11)/silica nanocomposites were prepared via in situ melt polymerization by the dispersion of hydrophobic silica in 11-aminoundecanoic acid monomer. Their isothermal crystallization process and melting behaviors were analyzed by differential scanning calorimetry. The isothermal crystallization kinetics was analyzed by the Avrami equation. The obtained data showed that the model of nucleation and growth of PA 11 was not affected after the incorporation of silica and was a mixture with two-dimensional, circular, three-dimensional growth with thermal nucleation. Double and single melting peaks were observed depending on the crystallization temperature. The equilibrium melting point of samples was evaluated, and the spherulites growth kinetics parameters and fold surface free energy were further calculated according to the classical theories. © 2011 Wiley Periodicals, Inc. J Appl Polym Sci, 2011 [ABSTRACT FROM AUTHOR]

The article discusses the findings of an experiment to predict the melting and crystallization behaviors, and morphology of pure polypropylene (PP) blends. It notes the use of the Avrami and Ozawa equations to analyze the nonisothermal crystallization kinetics of the PP blends, while surface fold free energy and activation energy were derived from the Hoffman-Lauritzen theory. The study attributes the reduced mobility and accelerated crystallization rate of PP chains to the presence of nylon 11.

Wind energy harvesting has attracted attention for the past few decades as a potential solution to supply sustainable power for wireless sensors network (WSN). In this paper, we reported an aeroacoustics-driven piezoelectric energy generator (APEG) without moving parts to scavenge wind energy. Aeroacoustics principle, a complex Multiphysics process including the coupling among physics fields of fluid, solid, acoustics and electricity, was used for converting wind energy to mechanical energy applied on the APEG. At 79.6 m/s jet velocity, the APEG delivers high closed-circuit current (4.7 mA) at 1.5 kΩ, open-circuit voltage (12.9 V), and large average output power (15.5 mW). This result was further validated by powering up different commercial electronic devices. Moreover, the complex aeroacoustics principle was studied both experimentally and theoretically. The velocity distribution of flow field and stress distribution of piezoelectric unit were simulated and analyzed. The proposed APEG with simple structure, small volume and high performance is promising for supplying sustainable power for WSN at harsh environment such as high-speed aircrafts. An aeroacoustics-driven piezoelectric energy generator (APEG) coupling with a quarter-wavelength resonator is proposed to scavenge wind energy for powering commercial electronic devices. The working mechanism is analyzed both experimentally and theoretically. The APEG exhibits high closed-circuit current (4.7 mA) and average output power (15.5 mW) under 1.5 kΩ. [Display omitted] • An aeroacoustics-driven piezoelectric energy generator (APEG) is developed for wind energy scavenging. • Aeroacoustics principle, the complex multiphysics process, is analyzed both experimentally and theoretically. • The velocity and stress distribution are simulated by finite element method (FEM) through COMSOL software. • Effect of piezoelectric layer thickness on RMS voltage/current and output power is analyzed. • The APEG is successfully demonstrated to power various commercial electronic devices. [ABSTRACT FROM AUTHOR]

In the new era of 5G and artificial intelligence of things (AIoT), remote monitoring and early warning based on widely distributed self-powered wireless sensing system are particularly important. However, the high power consumption, limited communication range and operation lifespan of this system are serious obstacles in practical applications. Here, we report a self-powered 5G narrowband IoT (NB-IoT) system for remote fire monitoring, driven by a magnetic-assisted noncontact energy harvester (MN-EH) which hybridizes piezoelectric and electromagnetic generator. The rotational MN-EH device can efficiently convert wind energy into electrical power while NB-IoT makes it possible to trade off the low-power and long-range transmission. With the magnetic coupling mechanism, the device can avoid degradation and abrasion to improve the reliability. Through optimizing a magnetic arrangement, the rotating peak torque can be reduced more than 93%. The MN-EH is further developed to power the NB-IoT system which can acquire temperature and CO concentration and transmit them to the cloud server every ~13 min. This work provides an effective strategy and paves the way for practical applications of self-powered remote monitoring system. A wind-driven fully self-powered 5G NB-IoT system is proposed to achieve remote fire monitoring and warning, driven by a magnetic-assisted noncontact energy harvester (MN-EH) which hybridizes piezoelectric and electromagnetic generator. [Display omitted] • A magnetic-assisted noncontact energy harvester (MN-EH) is developed for self-powered remote monitoring. • The optimized magnetic arrangement is utilized to reduce the rotating torque. • The MN-EH is successfully demonstrated to power a programmable 5G NB-IoT system. [ABSTRACT FROM AUTHOR]

Here, we subjected the marine copepod Tigriopus japonicus to environmentally-relevant concentrations of microplastics (MPs) and mercury (Hg) for three generations (F0–F2) to investigate their physiological and molecular responses. Hg accumulation and phenotypic traits were measured in each generation, with multi-omics analysis conducted in F2. The results showed that MPs insignificantly impacted the copepod's development and reproduction, however, which were significantly compromised by Hg exposure. Interestingly, MPs significantly increased Hg accumulation and consequently aggravated this metal toxicity in T. japonicus , demonstrating their carrier role. Multi-omics analysis indicated that Hg pollution produced numerous toxic events, e.g., induction of apoptosis, damage to cell/organ morphogenesis, and disordered energy metabolism, ultimately resulting in retarded development and decreased fecundity. Importantly, MPs enhanced Hg toxicity mainly via increased oxidative apoptosis, compromised cell/organ morphogenesis, and energy depletion. Additionally, phosphoproteomic analysis revealed extensive regulation of the above processes, and also impaired neuron activity under combined MPs and Hg exposure. These alterations adversely affected development and reproduction of T. japonicus. Overall, our findings should offer novel molecular insights into the response of T. japonicus to long-term exposure to MPs and Hg, with a particular emphasis on the carrier role of MPs on Hg toxicity. [Display omitted] • Copepod's multigenerational exposure to MPs and Hg was performed. • MPs insignificantly impacted the copepod's development and reproduction. • MPs increased Hg toxicity to the growth/reproduction in copepods. • Hg-induced compromised traits were linked to energy depletion and high apoptosis. • MPs enhanced Hg toxicity by high oxidative apoptosis and impaired neuron activity. [ABSTRACT FROM AUTHOR]

Background: Clear cell renal cell carcinoma (ccRCC) is a highly lethal malignancy with few therapeutic options. Cyclin‑dependent kinase 9 (CDK9), a potential therapeutic target of many cancers, has been recently observed to be upregulated in ccRCC patients. Therefore, we aimed to investigate the therapeutic potential of CDK9 in ccRCC and develop a novel CDK9 inhibitor with low toxicity for ccRCC treatment. Methods: The expression of CDK9 in ccRCC was checked using the online database and tissue microarray analysis. shRNA-mediated CDK9 knockdown and CDK inhibitor were applied to evaluate the effect of CDK9 on ccRCC. Medicinal chemistry methods were used to develop a new CDK9 inhibitor with drugability. RNA-seq and ChIP-seq experiments were conducted to explore the mechanism of action. MTS, western blotting, and colony formation assays were performed to evaluate the anti-ccRCC effects of CDK9 knockdown and inhibition in vitro. The in vivo anti-tumour efficacy was evaluated in a xenograft model. Results: CDK9 is overexpressed and associated with poor survival in ccRCC. Knockdown or inhibition of CDK9 significantly suppressed ccRCC cells. XPW1 was identified as a new potent and selective CDK9 inhibitor with excellent anti-ccRCC activity and low toxicity. In mechanism, XPW1 transcriptionally inhibited DNA repair programmes in ccRCC cells, resulting in an excellent anti-tumour effect. CDK9 and BRD4 were two highly correlated transcriptional regulators in ccRCC patients, and the BRD4 inhibitor JQ1 enhanced XPW1's anti-ccRCC effects in vitro and in vivo. Conclusions: This work provides valuable insights into the therapeutic potential of CDK9 in ccRCC. The CDK9 inhibitor XPW1 would be a novel therapeutic agent for targeting ccRCC, alone or in rational combinations. [ABSTRACT FROM AUTHOR]

One of the crucial challenge for developing high safety and high voltage lithium ion batteries is to find a reliable electrolyte system. In this work, we report a kind of hybrid electrolytes, which are used for high-voltage lithium ion batteries and are expected to be able to effectively enhance the battery safety. The hybrid electrolytes are obtained by incorporating silane-Al 2 O 3 (Al 2 O 3 -ST) into liquid electrolyte, which combines the merits of both solid electrolyte and liquid electrolyte. The Al 2 O 3 -ST nanoparticles help to increase lithium-ion transference number and to enhance battery safety, while liquid electrolyte contributes to high ionic conductivity. The cycling stability and rate capacity of LiNi 0.5 Mn 1.5 O 4 /Li batteries are improved by using the hybrid electrolytes. Nail-penetration tests indicate that LiNi 0.6 Mn 0.2 Co 0.2 O 2 /graphite battery with hybrid electrolyte owns obviously enhanced safety than that using traditional liquid electrolyte. This work provides new insight on electrolyte design for high-safety high-voltage lithium ion batteries. [ABSTRACT FROM AUTHOR]

In this study, polystyrene-b-poly (ethylene/propylene)-b-polystyrene grafted glycidyl methacrylate (SEPS-g-GMA) copolymer was prepared by melt grafting in the mixer. The successful grafting of GMA on the SEPS molecular chain was analyzed by Fourier transform infrared spectroscopy (FT-IR). The grafting ratio of SEPS-g-GMA was measured by acid-base titration, which showed that N-vinylpyrrolidone (NVP) had a positive effect on the increased grafting ratio of SEPS-g-GMA. The recycled polypropylene (R-PP)/recycled high impact polystyrene (R-HIPS) blends were prepared by a melt extrusion, and the effect of SEPS-g-GMA copolymer with different ratios on the compatibility of blends was studied. The epoxy groups in SEPS-g-GMA copolymer were found to have a chemical reaction with the carboxyl groups in the waste materials. When the grafting ratio of SEPS-g-GMA was 2.44%, the notched impact strength and the elongation at break of the R-PP/R-HIPS blends with 10 phr SEPS-g-GMA reached 7.06 kJ/m 2 and 34.25%, which were significantly increased by 186.99% and 68.30% compared with that of pure blends, respectively. Moreover, the observed decreased particle size and increased dispersion uniformity improved the compatibility using SEPS-g-GMA. The increased complex viscosity, storage modulus and loss modulus indicated that the chemical reaction between SEPS-g-GMA and R-PP/R-HIPS blends improved the component compatibility among the blends, resulting in the chain entanglement prominently. The presence of SEPS-g-GMA inhibited the degradation and increased the thermal stability of R-PP/R-HIPS blends. [ABSTRACT FROM AUTHOR]

Based on the CFD-DEM coupling method, this paper explores the dilute phase pneumatic conveying characteristics of shrimp feed particles. Furthermore, an experimental platform is established to verify the accuracy of the model. The simulated pipeline structure consists of a long horizontal pipe with horizontal bends and long vertical pipe with vertical bends. The results demonstrate a non-linear relationship between the pressure drop of the vertical pipeline and the inlet wind speed. Initially, the pressure drop decreases as the inlet wind speed increases, but it eventually starts to increase. When the inlet wind speed is not less than 18 m/s, the pressure drop of the horizontal pipeline increases with an accelerated inlet wind speed. The optimal inlet wind speed for adult shrimp farming is 24 m/s, while that for juvenile shrimp farming is 21 m/s. A correlation between system energy consumption and solid-gas ratio has been constructed. The ideal solid-gas ratio for adult shrimp farming is 3.69, while that for juvenile shrimp farming is 1.9–2.75. The solid-gas ratio and particle distribution at the inlet section of the straight pipe have a significant impact on the pressure drop characteristics. Their variation with the conveying wind speed directly influences the pressure drop patterns observed in the initial section of the straight pipe. This paper's findings offer valuable scientific and practical insights for the dilute phase pneumatic conveying of shrimp feed particles, providing guidance for effective and efficient processes in this domain. [Display omitted] • The best conveying wind speed with the lowest unit pressure drop is proposed. • The internal causes of pressure drop variations in long straight pipes are discussed. • The best solid gas ratio with the lowest system power consumption is given. [ABSTRACT FROM AUTHOR]

Abstract Traditional collaborative representation based classification (CRC) method usually faces the challenge of data uncertainty hence results in poor performance, especially in the presence of appearance variations in pose, expression and illumination. To overcome this issue, this paper presents a CRC-based face classification method by jointly using block weighted LBP and analysis dictionary learning. To this end, we first design a block weighted LBP histogram algorithm to form a set of local histogram-based feature vectors instead of using raw images. By this means we are able to effectively decrease data redundancy and uncertainty derived from image noises and appearance variations. Second, we adopt an analysis dictionary learning model as the projection transform to construct an analysis subspace, in which a new sample is characterized with the improved sparsity of its reconstruction coefficient vector. The crucial role of the analysis dictionary learning method in CRC is revealed by its capacity of the collaborative representation in an analytic coefficient space. Extensive experimental results conducted on a set of well-known face databases demonstrate the merits of the proposed method. [ABSTRACT FROM AUTHOR]

Recovering facial albedo from low quality face images is a challenging task which arises when face recognition is attempted in the wild. Low quality of facial images is usually caused by extrinsic factors such as low resolution and noises, and intrinsic ones such as expressions. Existing research recovers facial albedo by dealing with the extrinsic and intrinsic factors separately. However, it is more natural and potentially more useful to approach albedo recovery by removing the two effects simultaneously. In this paper, we present a novel framework which can recover facial albedo by jointly solving these for both the extrinsic and intrinsic sources of uncertainty. This framework models albedo recovery problem by a joint optimization process which alternatively (1) removes intra-personal variations and (2) performs super resolution. To deal with the intrinsic sources of albedo variability, we use a linear model. To handle extrinsic problems associated with low quality imaging, we use a sparse coding method which is applied to super resolution. The proposed method can also significantly improve the performance of face recognition and clustering in case of very low resolution and in the presence of various facial variations. Extensive experiments and comparisons are conducted on the AR and FERET face databases. Experimental results show the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

The STING-mediated type I interferon (IFN) signaling pathway has been shown to play critical roles in antitumor immunity. Here, we demonstrate that an endoplasmic reticulum (ER)-localized JmjC domain-containing protein, JMJD8, inhibits STING-induced type I IFN responses to promote immune evasion and breast tumorigenesis. Mechanistically, JMJD8 competes with TBK1 for binding with STING, blocking STING-TBK1 complex formation and restricting type I IFN and IFN-stimulated gene (ISG) expression as well as immune cell infiltration. JMJD8 knockdown improves the efficacy of chemotherapy and immune checkpoint therapy in treating both human and mouse breast cancer cell-derived implanted tumors. The clinical relevance is highlighted in that JMJD8 is highly expressed in human breast tumor samples, and its expression is inversely correlated with that of type I IFN and ISGs as well as immune cell infiltration. Overall, our study found that JMJD8 regulates type I IFN responses, and targeting JMJD8 triggers antitumor immunity. [Display omitted] • ER-localized protein JMJD8 inhibits STING-induced type I IFN signaling pathway • JMJD8 interacts with STING to disrupt the STING-TBK1 complex formation • JMJD8 knockdown inhibits breast tumorigenesis by activating antitumor immunity • JMJD8 knockdown improves chemotherapy and ICT effects in suppressing tumor growth Yi et al. demonstrate that ER-localized JMJD8 interacts with STING to disturb the STING-TBK1 complex formation, suppresses type I IFN responses, and promotes immune evasion in breast tumor. JMJD8 knockdown improves the efficacy of chemotherapy and immune checkpoint therapy in suppressing breast tumor growth. [ABSTRACT FROM AUTHOR]

Here, a novel three dimensional Ni foam-supported porous SnO 2 film (SnO 2 /Ni composite foam) was successfully synthesized by the simple hydrothermal method. The porous SnO 2 film had closely coated on the backbones of Ni foam without assistant of surfactant at mild conditions. At the same time, the pore size of SnO 2 /Ni composite foam was easily controlled by the reaction temperature. Furthermore, it was found that the capacitive properties of SnO 2 /Ni composite foam were related to the pore size of SnO 2 film. The SnO 2 film with smaller pore size showed higher capacitive properties, in which the porous SnO 2 /Ni composite foam with pore size of ca.60.0 nm showed high rate capability of 541.0 F g −1 at 10.0 A g −1 and good cycle stability with capacitance retention of 98.1% after 1000 cycles. The high capacitive properties was due to the SnO 2 active materials grown on the backbone of Ni foam and porous structure of SnO 2 materials, resulting in the reduce of resistance and enhancement of active surface area. Our work not only demonstrates the controlled synthesis of high-quality porous SnO 2 /Ni composite foam at mild conditions on a large scale, but also provides a universal route for the rational design of supercapacitors with high performance. [ABSTRACT FROM AUTHOR]

Basal-like breast cancer (BLBC) is the most aggressive subtype of breast cancer with a poor prognosis. Long noncoding RNAs (lncRNAs) play critical roles in human cancers. Krüppel-like Factor 5 (KLF5) is a key oncogenic transcription factor in BLBC. However, the underlying mechanism of mutual regulation between KLF5 and lncRNA remains largely unknown. Here, we demonstrate that lncRNA KPRT4 promotes BLBC cell proliferation in vitro and in vivo. Mechanistically, KLF5 directly binds to the promoter of KPRT4 to promote KPRT4 transcription. Reciprocally, KPRT4 recruits the YB-1 transcription factor to the KLF5 promoter by interacting with YB-1 at its 5' domain and forming an RNA-DNA-DNA triplex structure at its 3' domain, resulting in enhanced transcription of KLF5 and ultimately establishing a feedforward circuit to promote cell proliferation. Moreover, the antisense oligonucleotide (ASO)-based therapy targeting KPRT4 substantially attenuated tumor growth in vivo. Clinically, the expression levels of YB-1, KLF5 and KPRT4 are positively correlated in clinical breast specimens. Together, our data suggest that KPRT4 is a major molecule for BLBC progression and that the feedforward circuit between KLF5 and KPRT4 may represent a potential therapeutic target in BLBC. [ABSTRACT FROM AUTHOR]

Although the weakly supervised learning can effectively avoid the tedious data annotating process of deep learning approaches, the performance is still in urgent need of enhancement. In this paper, we endeavor to mine a ubiquitous and fundamental knowledge— Relation , to boost several existing classification and localization models without changing the original structure. We first propose a universal relation exploring scheme to mine the relations among entities. This scheme can be specialized into different instantiations including object, superpixel and pixel relations to stimulating different learning models. We adopt the object relations on a few-shot classification model to concentrate on the dominant object, and to boost its discriminative capacity. The superpixel relation is utilized to improve the performance of the saliency object detection models. The sensitivity of the pixel relations to the uncertain regions makes it suitable for distinguishing the disputed area in saliency detection results. Our experiments demonstrate that all the three relation instantiations can significantly boost the performance of the state-of-the-art learning models and optimize the visual result. [ABSTRACT FROM AUTHOR]

Real-time telemedicine detection can solve the problem of the shortage of public medical resources caused by the coming aging society. However, the development of such an integrated monitoring system is hampered by the need for high-performance sensors and the strict-requirement of long-distance signal transmission and reproduction. Here, a bionic crack-spring fiber sensor (CSFS) inspired by spider leg and cirrus whiskers for stretchable and weavable electronics is reported. Trans-scale conductive percolation networks of multilayer graphene around the surface of outer spring-like Polyethylene terephthalate (PET) fibers and printing Ag enable a high sensitivity of 28475.6 and broad sensing range over 250%. The electromechanical changes in different stretching stages are simulated by Comsol to explain the response mechanism. The CSFS is incorporated into the fabric and realized the human-machine interactions (HMIs) for robot control. Furthermore, the 5G Narrowband Internet of Things (NB-IoT) system is developed for human healthcare data collection, transmission, and reproduction together with the integration of the CSFS, illustrating the huge potential of the approach in human-machine communication interfaces and intelligent telemedicine rehabilitation and diagnosis monitoring., (© 2024 The Authors. Advanced Science published by Wiley‐VCH GmbH.)

Recently, deep metric learning (DML) has achieved great success. Some existing DML methods propose adaptive sample mining strategies, which learn to weight the samples, leading to interesting performance. However, these methods suffer from a small memory (e.g., one training batch), limiting their efficacy. In this work, we introduce a data-driven method, meta-mining strategy with semiglobal information (MMSI), to apply meta-learning to learn to weight samples during the whole training, leading to an adaptive mining strategy. To introduce richer information than one training batch only, we elaborately take advantage of the validation set of meta-learning by implicitly adding additional validation sample information to training. Furthermore, motivated by the latest self-supervised learning, we introduce a dictionary (memory) that maintains very large and diverse information. Together with the validation set, this dictionary presents much richer information to the training, leading to promising performance. In addition, we propose a new theoretical framework that can formulate pairwise and tripletwise metric learning loss functions in a unified framework. This framework brings new insights to society and facilitates us to generalize our MMSI to many existing DML methods. We conduct extensive experiments on three public datasets, CUB200-2011, Cars-196, and Stanford Online Products (SOP). Results show that our method can achieve the state of the art or very competitive performance. Our source codes have been made available at https://github.com/NUST-Machine-Intelligence-Laboratory/MMSI.

Data augmentation has been observed playing a crucial role in achieving better generalization in many machine learning tasks, especially in unsupervised domain adaptation (DA). It is particularly effective on visual object recognition tasks as images are high-dimensional with an enormous range of variations that can be simulated. Existing data augmentation techniques, however, are not explicitly designed to address the differences between different domains. Expert knowledge about the data is required, as well as manual efforts in finding the optimal parameters. In this article, we propose a novel domain-adaptive augmentation method by making use of a state-of-the-art style transfer method and domain discrepancy measurement. Specifically, we measure the discrepancy between source and target domains, and use it as a guide to augment the original source samples using style transferred source-to-target samples. The proposed domain-adaptive augmentation method is data and model agnostic that can be easily incorporated with state-of-the-art DA algorithms. We show empirically that, by using this domain-adaptive augmentation, we are able to gradually reduce the discrepancy between the source and target samples, and further boost the adaptation performance using different DA algorithms on three popular domain adaption datasets.

Modified asphalt with high content SBS is widely used in asphalt pavement due to its excellent high and low temperature performance. However, its anti-aging performance is insufficient. In order to improve the anti-aging performance of SBS modified asphalt, nano-ZnO, nano-TiO 2 , nano-SiO 2 and polyphosphoric acid (PPA) were added to high content (6.5 wt%) linear SBS modified asphalt as anti-aging agents in this study. Moreover, Dynamic Shear Rheometer (DSR), Fluorescence Microscope, and Fourier Transform Infrared Spectroscopy were employed to reveal the mechanism, through the investigation of the rheological and microscopic properties of modified asphalt before and after aging. The results showed that the influence of nanoparticles on the rutting resistance and fatigue resistance of high content SBS modified asphalt is weak, mainly because there is only weak physical interaction between nanoparticles and the SBS modifier, but no obvious chemical reaction. The significant cross-networking structure of high content SBS modified asphalt even has an adverse effect on the anti-aging performance of nano-modifiers. However, PPA obviously makes the cross-linked network structure of SBS modified asphalt more compact, and significantly improves the performance after short-term aging and long-term aging, mainly due to the chemical reaction between PPA and the active groups in SBS modified asphalt.

The presence of widely used sulfamethoxazole (SMX) in wastewater poses a threat to aquatic organisms and humans. Here, the responses of the emerging microalgal-bacterial granular sludge (MBGS) process in treating SMX-containing wastewater were investigated. The results indicated that 1, 5 and 10 mg/L SMX had little effect on the removals of organics and nutrients after an acclimation period of three to five days. SMX reduced intracellular glycogen content of MBGS, while the production of chlorophyll and extracellular polymeric substances tended to be promoted. Furthermore, the potential mechanisms on how MBGS adapted to SMX were deciphered to be the alterations of microbial community structure and function of MBGS. SMX might be degraded intracellularly into a carbon source for microbial metabolism and the SMX degraders were suspected to be Scenedesmaceae, Rhodocyclaceae and Burkholderiaceae. This study suggests that the MBGS process can handle SMX-containing wastewater, advancing knowledge on MBGS for antibiotics degradation., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2022 Elsevier Ltd. All rights reserved.)

The purpose of this paper is to promote the construction of digital healthy cities and improve the living standards of urban residents. Based on this, this paper analyzes the development of healthy cities, and studies community public health safety emergency management and nursing insurance service optimization methods for healthy urban environment construction. First, the concept of digital healthy urban environment construction is discussed. Then, the role of environmental health is discussed. Finally, two methods are designed to study the emergency management of public health safety and nursing insurance services in urban communities under the condition of environmental health. The results show that in the environmental health score of the city, the scores of X1 (the urban air quality excellent rate) and X6 (citizens' satisfaction with the environmental quality) were relatively low between 2016 and 2018, below 0.5 points. The scores for the remaining 3 years were relatively high, above 0.5. The scores of X2 (green coverage rate of built-up area), X3 (average grade sound effect of environmental noise in urban area), X4 (harmless treatment rate of domestic waste) and X5 (centralized treatment rate of domestic sewage) were relatively high from 2016 to 2018, above 0.5 points, and relatively low in the remaining 3 years, below 0.5 points. Meanwhile, residents are very satisfied with information collection and information management in public health and safety emergency management, and the number of very satisfied people is basically more than 40%. Satisfaction with resource allocation and privacy management is high, and the number of very satisfied people is basically above 30%. However, the satisfaction with risk perception and management measures is very low, and the number of very satisfied people is basically below 20%. It shows that the current construction of the community's public health and safety emergency management system is relatively poor in terms of X2-X5, while the development of X1 and X6 is relatively mature. The research not only provides a reference for the construction and improvement of a digital healthy city, but also contributes to the improvement of the healthy life of urban residents., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The handling editor B-JH declared a shared affiliation with the authors GH at the time of review., (Copyright © 2022 Hu, Wang, Jiang, Tian, Deng and Liu.)

MixUp is an effective data augmentation method to regularize deep neural networks via random linear interpolations between pairs of samples and their labels. It plays an important role in model regularization, semisupervised learning (SSL), and domain adaption. However, despite its empirical success, its deficiency of randomly mixing samples has poorly been studied. Since deep networks are capable of memorizing the entire data set, the corrupted samples generated by vanilla MixUp with a badly chosen interpolation policy will degrade the performance of networks. To overcome overfitting to corrupted samples, inspired by metalearning (learning to learn), we propose a novel technique of learning to a mixup in this work, namely, MetaMixUp. Unlike the vanilla MixUp that samples interpolation policy from a predefined distribution, this article introduces a metalearning-based online optimization approach to dynamically learn the interpolation policy in a data-adaptive way (learning to learn better). The validation set performance via metalearning captures the noisy degree, which provides optimal directions for interpolation policy learning. Furthermore, we adapt our method for pseudolabel-based SSL along with a refined pseudolabeling strategy. In our experiments, our method achieves better performance than vanilla MixUp and its variants under SL configuration. In particular, extensive experiments show that our MetaMixUp adapted SSL greatly outperforms MixUp and many state-of-the-art methods on CIFAR-10 and SVHN benchmarks under the SSL configuration.

Long non-coding RNAs (lncRNAs) play critical roles in tumorigenesis and progression of colorectal cancer (CRC). However, functions of most lncRNAs in CRC and their molecular mechanisms remain uncharacterized. Here we found that lncRNA ITGB8-AS1 was highly expressed in CRC. Knockdown of ITGB8-AS1 suppressed cell proliferation, colony formation, and tumor growth in CRC, suggesting oncogenic roles of ITGB8-AS1. Transcriptomic analysis followed by KEGG analysis revealed that focal adhesion signaling was the most significantly enriched pathway for genes positively regulated by ITGB8-AS1. Consistently, knockdown of ITGB8-AS1 attenuated the phosphorylation of SRC, ERK, and p38 MAPK. Mechanistically, ITGB8-AS1 could sponge miR-33b-5p and let-7c-5p/let-7d-5p to regulate the expression of integrin family genes ITGA3 and ITGB3, respectively, in the cytosol of cells. Targeting ITGB8-AS1 using antisense oligonucleotide (ASO) markedly reduced cell proliferation and tumor growth in CRC, indicating the therapeutic potential of ITGB8-AS1 in CRC. Furthermore, ITGB8-AS1 was easily detected in plasma of CRC patients, which was positively correlated with differentiation and TNM stage, as well as plasma levels of ITGA3 and ITGB3. In conclusion, ITGB8-AS1 functions as a competing endogenous RNA (ceRNA) to regulate cell proliferation and tumor growth of CRC via regulating focal adhesion signaling. Targeting ITGB8-AS1 is effective in suppressing CRC cell growth and tumor growth. Elevated plasma levels of ITGB8-AS1 were detected in advanced-stage CRC. Thus, ITGB8-AS1 could serve as a potential therapeutic target and circulating biomarker in CRC., Competing Interests: Declaration of interests The authors declare no competing interests., (Copyright © 2021 The Authors. Published by Elsevier Inc. All rights reserved.)

A large amount of training data is usually crucial for successful supervised learning. However, the task of providing training samples is often time-consuming, involving a considerable amount of tedious manual work. In addition, the amount of training data available is often limited. As an alternative, in this paper, we discuss how best to augment the available data for the application of automatic facial landmark detection. We propose the use of a 3D morphable face model to generate synthesized faces for a regression-based detector training. Benefiting from the large synthetic training data, the learned detector is shown to exhibit a better capability to detect the landmarks of a face with pose variations. Furthermore, the synthesized training data set provides accurate and consistent landmarks automatically as compared to the landmarks annotated manually, especially for occluded facial parts. The synthetic data and real data are from different domains; hence the detector trained using only synthesized faces does not generalize well to real faces. To deal with this problem, we propose a cascaded collaborative regression algorithm, which generates a cascaded shape updater that has the ability to overcome the difficulties caused by pose variations, as well as achieving better accuracy when applied to real faces. The training is based on a mix of synthetic and real image data with the mixing controlled by a dynamic mixture weighting schedule. Initially, the training uses heavily the synthetic data, as this can model the gross variations between the various poses. As the training proceeds, progressively more of the natural images are incorporated, as these can model finer detail. To improve the performance of the proposed algorithm further, we designed a dynamic multi-scale local feature extraction method, which captures more informative local features for detector training. An extensive evaluation on both controlled and uncontrolled face data sets demonstrates the merit of the proposed algorithm.