Ensemble learning is an important group of machine learning techniques that aim to enhance the nonlinearity and generalization ability of a learning system by aggregating multiple learners. We found ...that ensemble techniques show great potential for improving the performance of traditional hyperspectral target detection algorithms, while at present, there are few previous works have been done on this topic. To this end, we propose an Ensemble based Constrained Energy Minimization (E-CEM) detector for hyperspectral image target detection. Classical hyperspectral image target detection algorithms like Constrained Energy Minimization (CEM), matched filter (MF) and adaptive coherence/cosine estimator (ACE) are usually designed based on constrained least square regression methods or hypothesis testing methods with Gaussian distribution assumption. However, remote sensing hyperspectral data captured in a real-world environment usually shows strong nonlinearity and non-Gaussianity, which will lead to performance degradation of these classical detection algorithms. Although some hierarchical detection models are able to learn strong nonlinear discrimination of spectral data, due to the spectrum changes, these models usually suffer from the instability in detection tasks. The proposed E-CEM is designed based on the classical CEM detection algorithm. To improve both of the detection nonlinearity and generalization ability, the strategies of “cascaded detection”, “random averaging” and “multi-scale scanning” are specifically designed. Experiments on one synthetic hyperspectral image and two real hyperspectral images demonstrate the effectiveness of our method. E-CEM outperforms the traditional CEM detector and other state-of-the-art detection algorithms. Our code will be made publicly available.
State-of-the-art two-stage object detectors apply a classifier to a sparse set of object proposals, relying on region-wise features extracted by RoIPool or RoIAlign as inputs. The region-wise ...features, in spite of aligning well with the proposal locations, may still lack the crucial context information which is necessary for filtering out noisy background detections, as well as recognizing objects possessing no distinctive appearances. To address this issue, we present a simple but effective Hierarchical Context Embedding (HCE) framework, which can be applied as a plug-and-play component, to facilitate the classification ability of a series of region-based detectors by mining contextual cues. Specifically, to advance the recognition of context-dependent object categories, we propose an image-level categorical embedding module which leverages the holistic image-level context to learn object-level concepts. Then, novel RoI features are generated by exploiting hierarchically embedded context information beneath both whole images and interested regions, which are also complementary to conventional RoI features. Moreover, to make full use of our hierarchical contextual RoI features, we propose the early-and-late fusion strategies ( i.e. , feature fusion and confidence fusion), which can be combined to boost the classification accuracy of region-based detectors. Comprehensive experiments demonstrate that our HCE framework is flexible and generalizable, leading to significant and consistent improvements upon various region-based detectors, including FPN, Cascade R-CNN, Mask R-CNN and PA-FPN. With simple modification, our HCE framework can be conveniently adapted to fit the structure of one-stage detectors, and achieve improved performance for SSD, RetinaNet and EfficientDet.
Display omitted
Nanoformulations with advantages in drug delivery, safety and pharmacodynamics have been booming as a promising strategy for cancer therapy. However, the traditional nanocarrier still ...suffers from the low drug loading capacity, potential systematic toxicity, unclear metabolism, and other uncertainties. To overcome these issues, carrier-free nanodrugs with desirable bioactivity were developed rapidly and drawn considerable attention. Meanwhile, the multifunctional self-delivery nanoarcheticture fabricated by a simple and “green” method, has significant advantages in synergistic cancer therapy and inhibition of multidrug resistant (MDR). Till now, carrier-free nanoparticles for tumor theranostics, phototherapy, chemotherapy, diagnose and synergistic therapy, have made outstanding progress. In this review, we make an integrated and exhaustive overview of lately reports on carrier-free nanodrug delivery systems formed by several active agents. We summarize the self-assembly and modified strategies, with emphasis on application superiority of carrier-free nanocrystal, and give new insight into the establishment of ideal nanosystems for cancer treatment.
Dysregulated extravillous trophoblast invasion and proliferation are known to increase the risk of recurrent spontaneous abortion (RSA); however, the underlying mechanism remains unclear. Herein, in ...our retrospective observational case-control study we show that villous samples from RSA patients, compared to healthy controls, display reduced succinate dehydrogenase complex iron sulfur subunit (SDHB) DNA methylation, elevated SDHB expression, and reduced succinate levels, indicating that low succinate levels correlate with RSA. Moreover, we find high succinate levels in early pregnant women are correlated with successful embryo implantation. SDHB promoter methylation recruited MBD1 and excluded c-Fos, inactivating SDHB expression and causing intracellular succinate accumulation which mimicked hypoxia in extravillous trophoblasts cell lines JEG3 and HTR8 via the PHD2-VHL-HIF-1α pathway; however, low succinate levels reversed this effect and increased the risk of abortion in mouse model. This study reveals that abnormal metabolite levels inhibit extravillous trophoblast function and highlights an approach for RSA intervention.
Display omitted
An environmentally benign and efficient hydrothermal carbonization method is widely applied for the preparation of carbon-based adsorbents. However, the adsorption capacity toward ...anionic species would be influenced due to the negatively charged surface of the traditional hydrothermal carbonaceous materials; moreover most of the carbonaceous materials were in the form of powder which restricted the practical applications. Herein, amino-rich hydrothermal carbon-coated electrospun polyacrylonitrile fiber (PAN@NC) adsorbents were obtained through one-step hydrothermal carbonization approach assisted by diethylenetriamine using polyacrylonitrile (PAN) fibers as the templates, which showed highly efficient adsorption for anionic pollutants. The PAN@NC fibers were characterized in detail to confirm their structures and composition. The flexible and robust PAN@NC fiber membrane exhibited high adsorption capacity and good regeneration and recycling ability toward the anionic metal ion Cr(VI) and herbicide 2,4-dichlorophenoxyacetic acid (2,4-D). According to the Langmuir model, the adsorption behaviors showed monolayer adsorption capacities of 290.70mg/g and 164.47mg/g for Cr(VI) and 2,4-D, respectively, which were higher than that of many other adsorbents. Recycling study indicated that the removal efficiencies for both pollutants retained above 90% after five cycles. These findings demonstrate that PAN@NC fibers are promising adsorbents for the removal of anionic pollutants from wastewater solutions.
A multivariate linear regression model was proposed to achieve short period prediction of PM2.5 (fine particles with an aerodynamic diameter of 2.5 μm or less). The main parameters for the proposed ...model included data on aerosol optical depth (AOD) obtained through remote sensing, meteorological factors from ground monitoring (wind velocity, temperature, and relative humidity), and other gaseous pollutants (SO2, NO2, CO, and O3). Beijing City was selected as a typical region for the case study. Data on the aforementioned variables for the city throughout 2015 were used to construct two regression models, which were discriminated by annual and seasonal data, respectively. The results indicated that the regression model based on annual data had (R2 = 0.766) goodness-of-fit and (R2 = 0.875) cross-validity. However, the regression models based on seasonal data for spring and winter were more effective, achieving 0.852 and 0.874 goodness-of-fit, respectively. Model uncertainties were also given, with the view of laying the foundation for further study.
The North China Craton (NCC) is bounded by two Paleozoic accretionary arc terranes: the North Qinling terrane to the south and the Bainaimiao terrane to the north. The timing of arc accretion to the ...NCC and the architecture of the Bainaimiao arc remain unclear. During the building and accretion of the arcs along its margins, the NCC experienced a long sedimentary hiatus since the Ordovician, which ended with the deposition of bauxite-bearing sediments in the Late Carboniferous. In this paper we report the U–Pb and Hf isotopes of detrital zircons from the Late Carboniferous bauxite layer and use these data to constrain the tectonic evolution of the margin of the NCC. The detrital zircons yield a minimum U–Pb age of ca. 310Ma and a prominent age peak at ca. 450Ma. Zircon crystals with ages of ca. 330Ma and ca. 1900Ma are more common in the bauxite samples from the northern part of the NCC than in those from the central part. The εHf(t) values of the ca. 450Ma detrital zircon crystals of the bauxite samples from the NCC are similar to those of the contemporaneous detrital zircon crystals from the North Qinling arc terrane to the south, but different from those of the contemporaneous detrital zircon crystals from the Bainaimiao arc terrane to the north. The ca. 450Ma detrital zircon crystals in the ca. 310Ma bauxite deposits are therefore interpreted to have been derived from the North Qinling arc terrane. The source of the ca. 330Ma detrital zircon crystals of the bauxite deposits is interpreted to be the northern margin of the NCC, where intermediate-felsic plutons formed at ca. 330Ma are common. The results from this study support the interpretation that the Paleozoic continental arc terranes and their concomitant back-arc basins were developed along the margins of the NCC before ca. 450Ma, and the arc complexes were subsequently accreted to the craton in the Late Carboniferous. This was then followed by the formation of a walled continental basin within the NCC.
Display omitted
•Carboniferous bauxite deposits in the NCC have two separate provenances.•Marginal arcs most likely consolidated with NCC until the Late Carboniferous.•Carboniferous arc accretion changed the NCC into a continental basin.
The electrochemical properties of transition metal oxides strongly depend on the coordination environment of metal atoms. Nevertheless, the relationship between the coordination environment of metal ...atoms and electrochemical performance of metal oxides is unclear, while the strategy of adjusting the coordination environment of metal atoms is rare. Herein, the engineering of the coordination environment of Mn atoms in manganese dioxides (MnO2) by using a triethanolamine (TEA) complex‐induced method is reported. The detailed experimental characterizations and density functional theory calculations show that the optimized Mn coordination environment with oxygen deficiency and more corner‐shared Mn–Mn shells results in apparent electron dislocation and forms an effective built‐in electrical field. As a result, the obtained MnO2‐TEA sample exhibits a high conductivity and an excellent ion diffusion capacity, with a remarkable specific capacitance of 417.5 F g−1 at 1 A g−1. At the power density of 450.0 W kg−1, the fabricated asymmetric supercapacitor delivers the maximal energy density (57.4 Wh kg−1). This work not only provides an effective strategy of adjusting the coordination environment of metal atoms in metal oxides, but also presents a deeper understanding of the electronic structure dependent electrochemical performance of electrode materials.
A simple triethanolamine complex method can significantly enhance the capacitive performance of MnO2 by adjusting the coordination environment of the Mn atoms, in which oxygen deficiency and more corner‐shared Mn–Mn shells (Mn–Mn2) are introduced to delocalize electrons and form a built‐in electrical field in MnO2, realizing a high conductivity and an excellent ion diffusion capacity.