Search

The efficient and complete extraction of uranium from aqueous solutions is crucial for safeguarding human health from potential radiotoxicity and chemotoxicity. Herein, an ultrathin 2D metal–organic framework (MOF) nanosheet with cavity structures was elaborately constructed, based on a calix[4]arene ligand. The large molecular skeleton and cup-shaped feature of the calix[4]arene enabled the as-prepared MOFs with large layer separations, which can be readily delaminated into ultrathin single-layer (∼1.25 nm) nanosheets. The incorporation of permanent cavity structures to the MOF nanosheets can fully utilize their structural features of readily accessible adsorption groups and exposed surface area in uranium removal, reaching ultrafast adsorption kinetics; the functionalized cavity structure endowed MOF nanosheets with the ability to preconcentrate and extract uranium from aqueous solutions with ultrahigh efficiencies, even at extremely low concentrations. As a result, relatively high removal ratios (>95%) can be achieved for uranium within 5 min, even in the ultralow concentration range of 75–250 ppb, and the residual uranium was reduced to below 4.9 ppb. The MOF nanosheets also exhibited extremely high anti-interference ability, which could efficiently remove the low-level uranium (∼150 ppb) from various real samples. The characterizations and density functional theory calculations demonstrated that the synergistic effects of multiple interactions between the carboxylate groups and cage-like cavities with uranyl ions can be responsible for the efficient and selective uranium extraction.

This letter presents a novel millimeter-wave (mmWave) wideband ±45° dual-polarized (DP) phased array antenna (PAA) that incorporates heterogeneous beam elements (HBEs) to enable wide-angle scanning. The design of the wideband wide-beam ±45° DP HBE antenna encompasses intentionally crafted asymmetrical feeding structures and strategically positioned parasitic patches, enabling an asymmetrical aperture field and thus generating inclined beams. The proposed ±45° DP HBE antenna exhibits a wide impedance bandwidth, high polarization isolation, a wide 3 dB beamwidth, and significant beam inclination adjustability. Leveraging these HBEs, a 1 × 4-element wideband ±45° DP PAA is designed, fabricated, and measured. By utilizing HBEs with varying beam inclinations, the effective aperture of the PAA can be controlled at different scanning angles, facilitating wide-angle scanning. The measurement results confirm that the PAA offers an extensive scanning range, consistently achieving a 3 dB gain variation exceeding ±60° with a broadside gain of over 9.0 dB across the entire 24.0 GHz to 29.5 GHz band, surpassing previous designs in terms of scanning coverage.

Cardiopulmonary progenitor cells (CPPs) constitute a minor subpopulation of cells that are commonly associated with heart and lung morphogenesis during embryonic development but completely subside after birth. This fact offers the possibility for the treatment of pulmonary heart disease (PHD), in which the lung and heart are both damaged. A reliable source of CPPs is urgently needed. In this study, we reprogrammed human cardiac fibroblasts (HCFs) into CPP-like cells (or induced CPPs, iCPPs) and evaluated the therapeutic potential of iCPP-derived exosomes for acute lung injury (ALI). iCPPs were created in passage 3 primary HCFs by overexpressing GLI1, WNT2, ISL1 and TBX5 (GWIT). Exosomes were isolated from the culture medium of passage 6–8 GWIT-iCPPs. A mouse ALI model was established by intratracheal instillation of LPS. Four hours after LPS instillation, ALI mice were treated with GWIT-iCPP-derived exosomes (5 × 109, 5 × 1010particles/mL) via intratracheal instillation. We showed that GWIT-iCPPs could differentiate into cell lineages, such as cardiomyocyte-like cells, endothelial cells, smooth muscle cells and alveolar epithelial cells, in vitro. Transcription analysis revealed that GWIT-iCPPs have potential for heart and lung development. Intratracheal instillation of iCPP-derived exosomes dose-dependently alleviated LPS-induced ALI in mice by attenuating lung inflammation, promoting endothelial function and restoring capillary endothelial cells and the epithelial cells barrier. This study provides a potential new method for the prevention and treatment of cardiopulmonary injury, especially lung injury, and provides a new cell model for drug screening.

Multisubstituted furans occupy a pivotal position within the realms of synthetic chemistry and pharmacological science due to their distinctive chemical configurations and inherent properties. We herein introduce a tandem difunctionalization protocol of alcohols for the efficient synthesis of multisubstituted 2,3-dihydrofurans and γ-butyrolactones through the combination of photocatalysis and iron catalysis under mild conditions. Photoredox alcohol α-C(sp3)–H activation and Pinner-type intramolecular cyclization are two key processes. This method features significant convenience, economic benefits, and environmental friendliness.

Adult individuals with Down syndrome (DS) develop Alzheimer disease (AD). Whether there is a difference between AD in DS and AD regarding the structure of amyloid-β (Aβ) and tau filaments is unknown. Here we report the structure of Aβ and tau filaments from two DS brains. We found two Aβ40filaments (types IIIa and IIIb) that differ from those previously reported in sporadic AD and two types of Aβ42filaments (I and II) identical to those found in sporadic and familial AD. Tau filaments (paired helical filaments and straight filaments) were identical to those in AD, supporting the notion of a common mechanism through which amyloids trigger aggregation of tau. This knowledge is important for understanding AD in DS and assessing whether adults with DS could be included in AD clinical trials.

Repair of large bone defects caused by severe trauma, non-union fractures, or tumor resection remains challenging because of limited regenerative ability. Typically, these defects heal through mixed routines, including intramembranous ossification (IMO) and endochondral ossification (ECO), with ECO considered more efficient. Current strategies to promote large bone healing via ECO are unstable and require high-dose growth factors or complex cell therapy that cause side effects and raise expense while providing only limited benefit. Herein, we report a bio-integrated scaffold capable of initiating an early hypoxia microenvironment with controllable release of low-dose recombinant bone morphogenetic protein-2 (rhBMP-2), aiming to induce ECO-dominated repair. Specifically, we apply a mesoporous structure to accelerate iron chelation, this promoting early chondrogenesis via deferoxamine (DFO)-induced hypoxia-inducible factor-1α (HIF-1α). Through the delicate segmentation of click-crosslinked PEGylated Poly (glycerol sebacate) (PEGS) layers, we achieve programmed release of low-dose rhBMP-2, which can facilitate cartilage-to-bone transformation while reducing side effect risks. We demonstrate this system can strengthen the ECO healing and convert mixed or mixed or IMO-guided routes to ECO-dominated approach in large-size models with clinical relevance. Collectively, these findings demonstrate a biomaterial-based strategy for driving ECO-dominated healing, paving a promising pave towards its clinical use in addressing large bone defects.

Two novel MoO42–-templated luminescent silver alkynyl nanoclusters with 20-nuclearity ([(MoO42–)@Ag20(C=CtBu)8(Ph2PO2)7(tfa)2]·(tfa–) (1)) and 18-nuclearity ([(MoO42–)@Ag18(C=CtBu)8(Ph2PO2)7]·(OH) (2)) (tfa = trifluoroacetate) were synthesized with the green light maximum emissions at 507 and 516 nm, respectively. The nanoclusters were investigated and characterized by single-crystal X-ray crystallography, electrospray ionization mass spectrum (ESI-MS), X-ray photoelectron spectroscopy, thermogravimetry (TG), photoluminescence (PL), ultraviolet–visible (UV–vis) spectroscopy, and density functional theory calculations (DFT). The two nanoclusters differ in their structure by a supplementary [Ag2(tfa)2] organometallic surface motif, which significantly participates in the frontier molecular orbitals of 1, resulting in similar bonding patterns but different optical properties between the two clusters. Indeed, both nanoclusters show strong temperature-dependent photoluminescence properties, which make them potential candidates in the fields of optical devices for further applications.

Objective:This study investigates the application of metagenomic next-generation sequencing (mNGS) in the diagnosis of neurobrucellosis (NB). Methods:We retrospectively analyzed patients diagnosed with NB who underwent cerebrospinal fluid (CSF) mNGS testing in Xijing Hospital from 2015 to 2021. Results:Among the 20 individuals included in the study, the serum rose bengal test was positive in 11 out of 16 cases, serum agglutination test was positive in 13 out of 16 cases, CSF culture was positive in 6 out of 11 cases, and CSF mNGS tests were positive in 18 out of 20 cases. Conclusion:CSF mNGS demonstrates superior sensitivity; therefore, it is recommended to collect CSF for mNGS testing prior to antibiotic therapy when NB is suspected.

A successful synthesis of helical-shaped axially chiral bisoxime ethers is reported. This approach utilized symmetric L-shaped diketone scaffolds as carbonyl components for the enantioselective condensation with hydroxylamines, delivering dual axially chiral oxime ethers with up to 99% ee. Additionally, the axially chiral mono-oxime ethers of azabicyclic ketones with high ee’s were also successfully produced. Various chiral bicyclic lactams can be readily synthesized via Beckmann rearrangement, demonstrating a potential application in organic synthetic chemistry.

The complexity and heterogeneity of individual tumors have hindered the efficacy of existing therapeutic cancer vaccines, sparking intensive interest in the development of more effective in situvaccines. Herein, we introduce a cancer nanovaccine for reactive oxygen species-augmented metalloimmunotherapy in which FeAl-layered double hydroxide (LDH) is used as a delivery vehicle with dihydroartemisinin (DHA) as cargo. The LDH framework is acid-labile and can be degraded in the tumor microenvironment, releasing iron ions, aluminum ions, and DHA. The iron ions contribute to aggravated intratumoral oxidative stress injury by the synergistic Fenton reaction and DHA activation, causing apoptosis, ferroptosis, and immunogenic cell death in cancer cells. The subsequently released tumor-associated antigens with the aluminum adjuvant form a cancer nanovaccine to generate robust and long-term immune responses against cancer recurrence and metastasis. Moreover, Fe ion-enabled T1-weighted magnetic resonance imaging can facilitate real-time tumor therapy monitoring. This cancer-nanovaccine-mediated metalloimmunotherapy strategy has the potential for revolutionizing the precision immunotherapy landscape.

Calcium nanoparticles have been investigated for applications, such as drug and gene delivery. Additionally, Ca2+serves as a crucial second messenger in the activation of immune cells. However, few studies have systematically studied the effects of calcium nanoparticles on the calcium levels and functions within immune cells. In this study, we explore the potential of calcium nanoparticles as a vehicle to deliver calcium into the cytosol of dendritic cells (DCs) and influence their functions. We synthesized calcium hydroxide nanoparticles, coated them with a layer of silica to prevent rapid degradation, and further conjugated them with anti-CD205 antibodies to achieve targeted delivery to DCs. Our results indicate that these nanoparticles can efficiently enter DCs and release calcium ions in a controlled manner. This elevation in cytosolic calcium activates both the NFAT and NF-κB pathways, in turn promoting the expression of costimulatory molecules, antigen-presenting molecules, and pro-inflammatory cytokines. In mouse tumor models, the calcium nanoparticles enhanced the antitumor immune response and augmented the efficacy of both radiotherapy and chemotherapy without introducing additional toxicity. Our study introduces a safe nanoparticle immunomodulator with potential widespread applications in cancer therapy.

A method for preparing high-purity tellurium based on viscous distillation and gas-phase separation was developed. By adjusting the condenser structure and condensing temperature to achieve viscous distillation, the dominant condensation zone for tellurium was controlled; hence, poorly volatile substances remained in the crucible, while selenium was separated from tellurium in the gas phase to achieve the purification of metallurgical-grade crude tellurium. Specifically, the saturated vapor pressure, mean free path of tellurium, impurity distribution, and morphologies of the condensate at different locations were investigated in detail. The results for a 500 g upscaled experiment showed that the direct yield of Te was 83.40%, the purity of Te reached higher than 99.9996 wt.% and the selenium content in the dominant condensation zone was reduced to 2.76×10−4wt.% under the optimal experimental conditions of 5 Pa, 773 K and 90 min.

The current live rotavirus (RV) vaccines show reduced effectiveness in developing countries, calling for vaccine strategies with improved efficacy and safety. We generated pseudovirus nanoparticles (PVNPs) that display multiple ectodomains of RV viral protein 4 (VP4), named S-VP4e, as a nonreplicating RV vaccine candidate. The RV spike protein VP4s that bind host receptors and facilitate viral entry are excellent targets for vaccination. In this study, we developed scalable methods to produce three S-VP4e PVNPs, each displaying the VP4e antigens from one of the three predominant P[8], P[4], and P[6] human RVs (HRVs). These PVNPs were recognized by selected neutralizing VP4-specific monoclonal antibodies, bound glycan receptors, attached to permissive HT-29 cells, and underwent cleavage by trypsin between VP8* and VP5*. 3D PVNP models were constructed to understand their structural features. A trivalent PVNP vaccine containing the three S-VP4e PVNPs elicited high and well-balanced VP4e-specific antibody titers in mice directed against the three predominant HRV P types. The resulting antisera neutralized the three HRV prototypes at high titers; greater than 4-fold higher than the neutralizing responses induced by a trivalent vaccine consisting of the S60-VP8* PVNPs. Finally, the trivalent S-VP4e PVNP vaccine provided 90–100% protection against diarrhea caused by HRV challenge. Our data supports the trivalent S-VP4e PVNPs as a promising nonreplicating HRV vaccine candidate for parenteral delivery to circumvent the suboptimal immunization issues of all present live HRV vaccines. The established PVNP-permissive cell and PVNP-glycan binding assays will be instrumental for further investigating HRV–host cell interactions and neutralizing effects of VP4-specific antibodies and antivirals.

[Display omitted] Methanol steam reforming (MSR) is a promising way of producing hydrogen. The catalyst is a key factor in the MSR reaction. This review mainly describes the latest progress in MSR catalysts, including the development of catalysts, reaction mechanisms, and kinetic analysis, to elucidate the design of efficient catalyst systems and the basic understanding of methanol conversion at the molecular level. MSR catalysts are mainly divided into three categories: copper-based catalysts, precious metal catalysts, and molybdenum carbide-based catalysts. In each section, strategies for improving catalyst performance are described. The deactivation of catalysts is usually attributed to the formation of coke, sintering or oxidation of active metals, and surface passivation. The reaction mechanism mainly consists of two types: methyl formate intermediates and formaldehyde intermediates. The insights of this review contribute to the further development of efficient catalysts and create value for methanol as a hydrogen carrier for fuel cell applications. [ABSTRACT FROM AUTHOR]

A high-efficiency dual-polarized conformal transmitarray antenna is proposed by using cross-arranged Huygens elements composed of electric and magnetic resonance structures in this letter. In order to achieve high aperture efficiency (AE), the Huygens elements with high transmittance and angular stability are used as the phase modulation surface of the antenna. The dual-polarization performance of the antenna is achieved by cross arrangement of the units. The effects of different curvatures on the two polarizations are investigated separately, and the simulation results prove that the antenna has good curvature stability. Finally, the proposed antenna is fabricated and measured. The thickness of the array is 9 mm (0.24λ at 8 GHz), and the radius of curvature is 300 mm. The highest gains are achieved at 7.8 GHz with 23.62 and 24.57 dBi for x-polarization and y-polarization, and the corresponding AEs are 41.66% (x-polarization) and 46.25% (y-polarization), respectively. The proposed transmitarray antenna can be independently controlled in two polarizations with high AE, and it has high practical value for airborne and satellite-based detection radars.

This article presents a speed-senserless control strategy with an adaptive speed observer using power winding (PW) information for dual-stator winding induction starter/generator, in order to improve speed estimation precision and starting-generating performances. Based on the principle of model reference adaptive system, an improved rotor-flux-based reference model is proposed by utilizing the electromagnetic link between the control winding and PW, the PW voltages are used to estimate rotor flux without the necessity of stator resistance and filter parameters. The improved speed observer is less parameter-sensitive and can reduce the estimation error, especially in a low-speed starting process with the load. The parameter design method of the adaptive mechanism of the proposed speed observer based on stability analysis is also given. The distribution of dominant pole of the speed observer is modified to obtain good tracking ability. Based on the improved speed observer, a speed-sensorless starting-generating control strategy is investigated, in which the estimated speed is not only used in speed control in starting process but also in staring-to-generating switching process. The simulation and experimental results verify the validity of the proposed speed observer and the speed-sensorless control strategy.

Nanomedicines have been approved to treat multiple human diseases. However, clinical adoption of nanoformulated agents is often hindered by concerns about hepatic uptake and clearance, a process that is not fully understood. Here we show that the antitumour efficacy of cancer nanomedicine exhibits an age-associated disparity. Tumour delivery and treatment outcomes are superior in old versus young mice, probably due to an age-related decline in the ability of hepatic phagocytes to take up and remove nanoparticles. Transcriptomic- and protein-level analysis at the single-cell and bulk levels reveals an age-associated decrease in the numbers of hepatic macrophages that express the scavenger receptor MARCO in mice, non-human primates and humans. Therapeutic blockade of MARCO is shown to decrease the phagocytic uptake of nanoparticles and improve the antitumour effect of clinically approved cancer nanotherapeutics in young but not aged mice. Together, these results reveal an age-associated disparity in the phagocytic clearance of nanotherapeutics that affects their antitumour response, thus providing a strong rationale for an age-appropriate approach to cancer nanomedicine.

In this work, a new treatment method combining ultrasonic vibration with FeCoNiCrCu high entropy alloy (HEA) coating was used to prepared Al/Mg bimetal through the lost foam compound casting. The effects of composite treatment involving ultrasonic vibration and HEA coating on interfacial microstructure and mechanical properties of Al/Mg bimetal were studied. Results demonstrate that the interface thickness of the Al/Mg bimetal with composite treatment significantly decreases to only 26.99% of the thickness observed in the untreated Al/Mg bimetal. The HEA coating hinders the diffusion between Al and Mg, resulting the significant reduction in Al/Mg intermetallic compounds in the interface. The Al/Mg bimetal interface with composite treatment is composed of Al3Mg2and Mg2Si/AlxFeCoNiCrCu+FeCoNiCrCu/δ-Mg+Al12Mg17eutectic structures. The interface resulting from the composite treatment has a lower hardness than that without treatment. The acoustic cavitation and acoustic streaming effects generated by ultrasonic vibration promote the diffusion of Al elements within the HEA coating, resulting in a significant improvement in the metallurgical bonding quality on the Mg side. The fracture position shifts from the Mg side of the Al/Mg bimetal only with HEA coating to the Al side with composite treatment. The shear strength of the Al/Mg bimetal increases from 32.16 MPa without treatment to 63.44 MPa with ultrasonic vibration and HEA coating, increasing by 97.26%.

Space-time-coding digital metasurface has drawn worldwide attention with the ability to improve communication quality and change the direction of electromagnetic (EM) wave propagation in real time. This article proposes a deep-learning-assisted method to design the space-time-coding digital metasurface element (STCDME) with the state recognition and mapping technique methods. Compared to traditional pure EM simulation methods to simulate all states, the proposed method fully considers the relationship between different states of STCDME to accelerate the design. First, we use the state recognition method to distinguish the S-parameters’ states. Subsequently, with the state mapping method, we use one state phase to predict the other states’ S-parameters. Simulation time is reduced by half with the proposed methods. Various algorithms are compared, and finally, the convolutional neural network (CNN)-long short-term memory (LSTM)-deep neural network (DNN) (CLD) hybrid algorithm is chosen for the proposed methods, which is also the first instance of using the CLD algorithm to design the EM structure. The proposed method is validated with two design examples. An element prototype is made and measured with the vector network analyzer (VNA). The measurement results agree with the simulation results. We then use the designed STCDME to complete the design of the beamforming array, which realizes the beamforming at 0°, 15°, 30°, and 45° angles. Finally, we briefly discuss the application/drawbacks of the proposed methods and the CLD model in EM fields.

Semantic segmentation is a significant task for remote sensing interpretation, which takes advantage of contextual semantic information to classify each pixel into a specific category. Most current methods apply convolutional neural networks (CNN) to learn feature representation from remote sensing images, which may ignore the global dependencies due to the limitation of convolutional kernels. Inspired by the global feature learning ability of Transformer, we propose a novel deep model called dual-path feature aware network (DPFANet), which combines the structure of CNN and Transformer for semantic segmentation of remote sensing images. DPFANet aims to learn effective modeling ability from local to global features of images. Simultaneously, an adaptive feature fusion network is developed to fuse features from dual-path networks. Moreover, an edge optimization block is applied to constrain the edge features, whose purpose is to obtain more representative features for segmentation. Experimental results on three public remote sensing datasets verify that our proposed network yields better segmentation performance compared to other related methods.

Graphical Abstract:

In heterogeneous image change detection (HICD), preventing neural networks from distorting critical information is the main challenge of such methods based on deep translation. Most of these methods rely on a priori information to suppress the effects of changed pixels in the translation process, but the accuracy of the prior information will influence the results of translation. In this article, we propose an end-to-end multidomain constrained translation network (MDCTNet) for unsupervised HICD. The proposed MDCTNet utilizes an improved generative adversarial network (GAN) to generate target domain images realistically. Furthermore, to retain the content information of the source domain images, MDCTNet leverages contrastive learning to ensure the consistency of adjacent pixel relationships. Meanwhile, it employs high-frequency information consistency which preserves pivotal characteristics. We compare the proposed MDCTNet with state-of-the-art algorithms to verify the efficacy of the proposed technique. The experimental results on five real datasets demonstrate the effectiveness of the proposed method.

How to improve emergency management is still an open issue. In real application, since it is unpractical to optimize all of influential factors, a feasible way is to find out the critical success factors (CSF) to improve. In this paper, the existing evidential DEMATEL method is improved to be more reasonable. Inspired by belief entropy, a new function which is used to calculate the reliability of the information is defined. Then, DEMATEL method is applied on each fused BPA multiplied by the reliability coefficient to seek for a final result. Finally, five critical success factors are figured out. By optimizing these five factors, the effectiveness and efficiency of the whole emergency management system could be greatly promoted.

A discrete-time sensorless current control method is proposed for the interior permanent magnet synchronous motor (IPMSM) in this article. Due to the inconsistent $dq$ -axis inductance of IPMSMs, the design of the discrete-time current control method and sensorless control method is complicated. The control methods are therefore widely designed in the continuous-time domain, while the performance of these continuous-time methods is poor. In the proposed scheme, the current regulator is designed in the discrete-frequency domain based on an entirely discretized state-space model, which improves the dynamic and steady-state performance of the current control system. In terms of the electrical angle estimation method, the continuous-time model-based method leads to a significant estimation error. To improve the accuracy of the estimated back-EMF, a state-space model with a modified back-EMF (MBEMF) model is derived in the discrete-time domain. Then the observer of MBEMF and a phase-locked loop (PLL) with phase compensation are proposed based on the discrete-time state-space model. The proposal improves the accuracy of the estimated electrical angle. The proposal is designed directly in the discrete-time domain, which avoids significant performance deterioration in digital implementations. Finally, the proposed method is implemented on an IPMSM drive test rig to verify its improvement.

Extracting users’ energy consumption patterns (ECPs) from smart meter data is an important work for retailers. The existing literature usually describe these patterns by clustering the daily load curves (DLCs), but lack a clear and quantified representation to explain what the exact schema of a user is. Therefore, this article proposes a new binary encoding method for ECPs quantification. Specifically, first, both time and value intervals are divided for dimensionality reduction based on the similarity of adjacent timestamp loads. Then, a binary aggregate approximation (BAX) method is proposed to encode each DLC into a BAX word, and the BAX words of users are merged to obtain the schemas with a three-element alphabet. Finally, based on the schemas, the stability scores of users’ patterns are quantified and are used to select target users for demand response (DR) measures. Case studies on a real dataset with 5566 users show that each target user averagely contributes to 0.172% of peak reduction, while each unselected user only contributes to 0.026%. Furthermore, to obtain target schemas and to find new users in future DR measures, a $K$ -means symbolic algorithm is designed to cluster BAX words of target users. The proposed encoding method and the findings can provide guidance of finding typical target users for DR measures.

The fusion of hyperspectral image (HSI) and light detection and ranging (LiDAR) data is widely used for land cover classification. However, due to different imaging mechanisms, HSI and LiDAR data always present significant image differences, and the dimensions and feature distributions of HSI and LiDAR are highly dissimilar. This makes it challenging to represent and correlate semantic information from multimodal data. Current methods for classifying pixel-by-pixel features, which rely on cascaded or attention-based fusion, cannot effectively use multimodal features. To achieve accurate classification results, extracting and fusing similar high-order semantic information and complementary discriminative information contained in multimodal data is vital. In this article, we propose a cross-modal semantic enhancement network (CMSE) for multimodal semantic information mining and fusion. Our proposed CMSE framework extracts features from the image on multiple scales, capturing more representative local sparse features with different sizes of convolution kernels. To represent high-level semantic features related to land cover, we establish a Gaussian-weighted matrix and semantically transform the spatial and spectral features of distinct branches. Finally, we build a multilevel residual fusion module to incrementally fuse spectral features from HSI and elevation features from LiDAR. Additionally, we introduce a cross-modal semantically constrained loss to guide multimodal semantic feature alignment. We evaluate our approach on three multimodal remote sensing (RS) datasets, namely the Houston2013, Trento, and MUUFL datasets. The experimental results demonstrate that our proposed CMSE model achieves superior performance in terms of accuracy and robustness compared to other related deep networks.

The pixel-level classification of multimodal remote sensing (RS) images plays a crucial role in the intelligent interpretation of RS data. However, existing methods that mainly focus on feature interaction and fusion fail to address the challenges posed by confounders—brought by sensor imaging bias, limiting their performance. In this article, we introduce causal inference into intelligent interpretation of RS and propose a new texture-aware causal feature extraction network (TeACFNet) for pixel-level fusion classification. Specifically, we propose a two-stage causal feature extraction (CFE) framework that helps networks learn more explicit class representations by capturing the causal relationships between multimodal heterogeneous data. In addition, to solve the problem of low-resolution land cover feature representation in RS images, we propose the refined statistical texture extraction (ReSTE) module. This module integrates the semantics of statistical textures in shallow feature maps through feature refinement, quantization, and encoding. Extensive experiments on two publicly available datasets with different modalities, including Houston2013 and Berlin datasets, demonstrate the remarkable effectiveness of our proposed method, which reaches a new state-of-the-art.

Multitarget domain adaptation (MTDA) presents a formidable challenge in remote sensing image scene classification (RSICS), where the objective is to transfer knowledge from a labeled source domain to several unlabeled target domains. Compared to single-source-single-target domain adaptation (S3TDA), MTDA is inherently more complex due to domain shifts among multiple target domains. Directly merging the unique features of multitarget domains can result in corrupted information and poor classification performance. To address these challenges, we propose a hierarchical feature progressive alignment network (HFPAN) for RSICS in MTDA. First, our method introduces a fine-grained and contextual information extraction (FCIE) network to extract the global-local correlation in remote sensing (RS) images. Second, we construct a hierarchical feature embedding (HFE) framework that maintains hierarchical inter, intra constraints for the extracted features. Finally, we perform an alignment process for the constructed hierarchical features to minimize the differences in MTDA, progressing from coarse to fine granularity. To evaluate the efficacy of our proposed method, we conducted several cross-domain scene classification experiments on five public datasets. These experiments demonstrate the novelty of our approach and its ability to achieve improved classification performance.

With the development of smart grid and renewable energy technologies, residential load forecasting has become an increasingly important task. Short-term residential load forecasting is not only conducive to power dispatching and peak-shaving and valley filling of the grid, but also good for residents to obtain higher economic benefit from renewable energy. This paper proposes a new baseline-refinement forecasting framework consisting of two main steps, baseline profile construction and refinement predictions. Firstly, a baseline profile construction method is proposed to forecast the baseline load profile based on the similarity and cyclic patterns of daily load profiles. Secondly, for the refinement predictions, a bi-attention mechanism is proposed by combining self-attention mechanism and external attention mechanism to fulfill the feature transformation and is included in the temporal convolutional network to refine the baseline profile. The final load forecasting results are obtained by aggregating the baseline profile and the refinement predictions. The simulation results demonstrate that the proposed framework has smaller forecasting errors and higher forecasting stability than the commonly used models on two load forecasting metrics.

This paper studies the spatial-temporal scheduling for electric vehicles with stochastic charging behaviors. A new spatial-temporal scheduling model is proposed to achieve both the spatial coordination and temporal coordination for EV users and is modeled as a non-convex optimization problem which can be solved by the coordinate descent method. Further, a new billing method of charging cost is proposed based on the Vickrey-Clarke-Groves to avoid possible inauthentic information from users. Specially, the individual rationality and incentive compatibility of the method are theoretically proved. Moreover, a stochastic optimization problem is proposed to describe the stochastic charging behavior of users, which can fit the real application better than the deterministic one. Consequently, an efficient charging strategy is obtained via chance constraint method. Simulation results and comparison studies show the effectiveness of both the model and the billing method.

This article introduces a machine learning (ML) framework for the design of space–time-coding digital metasurface elements (STCDMEs), commonly used in reconfigurable intelligent surface (RIS)-based communication. It includes inverse design, forward design, and automodeling, which quickly achieve multistate electromagnetic (EM) structure designs, e.g., STCDME. The decision tree (DT) model is chosen for use with its lightweight, fast response, and highly accurate in EM structure small-scale data modeling. In addition, we present a new sensing STCDME design method, using gap technology based on ground structure. Using the proposed framework, we successfully design a sensing STCDME with a reflection phase within 180° ± 30° ranging from 6.66 to 7.3 GHz and a reflection coefficient larger than −2 dB, meeting RIS communication requirements. In the 8.27–9.5-GHz band, the structure’s transmission coefficient exceeds −3 dB, achieving EM wave transmission and sensing capabilities. The proposed framework offers a novel method for STCDME design, and the resulting sensing STCDME structure can be used for RIS sensing, contributing significantly to wireless communication and sensing applications.

Purpose: The purposes of this study were to assess the changes in body composition in patients who underwent thyroidectomy due to differentiated thyroid cancer (DTC) after radioactive iodine therapy (RAI) and short-term levothyroxine (LT4) supplementation and to explore the correlations between body composition distribution and corresponding blood indices. Methods: Fifty-seven thyroidectomized DTC patients were included. Serum was tested for several biochemical indices of thyroid function, lipids, and bone metabolism, and body composition parameters were measured via dual-energy X-ray absorptiometry before and 4–6 weeks after RAI and LT4 supplementation. Results: The body composition of DTC patients changed after RAI. Fat mass in all parts of the body decreased (range of relative change (RRC) -12.97–-2.80%). Bone mineral content (BMC) increased throughout the body (relative change (RC) 12.12%), head (RC 36.23%), pelvis (RC 9.00%), and legs (RC 3.15%). Similarly, bone mineral density (BMD) increased in different regions (RRC 3.60–26.43%), except for the arms. Notably, lean mass in the arms (RC 4.30%) and legs (RC 3.67%) increased, while that in the head decreased (RC -2.75%), while total lean mass did not change at 4–6 weeks after LT4 supplementation. Furthermore, changes in fat distribution in the android region were related to the changes in total cholesterol (r= -0.390) and low-density lipoprotein cholesterol (r= -0.354), and changes in the BMC and BMD of the lumbar spine were positively associated with the changes in calcitonin (r= 0.302 and 0.325, respectively). Conclusions: After RAI and short-term LT4 supplementation in DTC patients, body composition rapidly and positively changed and was characterized by decreased fat mass and increased BMC and BMD.

Deep learning (DL) plays a vital role in the design of electromagnetic (EM) structures. However, in current research, a single neural network typically supports only one structure design and requires a complex framework to accommodate multiple structure designs. This article proposes using one neural-assisted design for facilitating multiple EM structures. We use two filling methods to control the vector length, an identification method to ensure accurate prediction results, and a random auxiliary vectors method to increase the data volume and reduce loss. Subsequently, we design a forward neural network (FNN) and an inverse neural network (INN) using the proposed method. The developed neural network is used to complete the dual-passband frequency-selective surface (DP-FSS), space–time-coding digital metasurface element (STCDME), single-/dual-band absorbing metasurface (SDAM), and dual-stopband frequency-selective surface (DS-FSS) designs. The mean absolute error (MAE) loss values for FNN/INN predictions and actual results for these four structures are 0.019/0.116, 0.035/0.602, 5.14/0.146, and 0.018/0.07, respectively. Finally, we design four structures with the well-training network, fabricate DP-FSS and DS-FSS, and measure them in an anechoic chamber. The measurement and simulation results are in good agreement. The proposed method significantly reduces the complexity of multiple EM structure designs, decreases the need for multiple neural networks, and simplifies the design framework, thereby contributing to the development of AI-assisted EM structures.