Search

With the development of deep learning technology and the widespread application of drones in the agricultural sector, the use of computer vision technology for target detection of pineapples has gradually been recognized as one of the key methods for estimating pineapple yield. When images of pineapple fields are captured by drones, the fruits are often obscured by the pineapple leaf crowns due to their appearance and planting characteristics. Additionally, the background in pineapple fields is relatively complex, and current mainstream target detection algorithms are known to perform poorly in detecting small targets under occlusion conditions in such complex backgrounds. To address these issues, an improved YOLOv8 target detection algorithm, named DPD-YOLO (Dense-Pineapple-Detection YOU Only Look Once), has been proposed for the detection of pineapples in complex environments. The DPD-YOLO model is based on YOLOv8 and introduces the attention mechanism (Coordinate Attention) to enhance the network’s ability to extract features of pineapples in complex backgrounds. Furthermore, the small target detection layer has been fused with BiFPN (Bi-directional Feature Pyramid Network) to strengthen the integration of multi-scale features and enrich the extraction of semantic features. At the same time, the original YOLOv8 detection head has been replaced by the RT-DETR detection head, which incorporates Cross-Attention and Self-Attention mechanisms that improve the model’s detection accuracy. Additionally, Focaler-IoU has been employed to improve CIoU, allowing the network to focus more on small targets. Finally, high-resolution images of the pineapple fields were captured using drones to create a dataset, and extensive experiments were conducted. The results indicate that, compared to existing mainstream target detection models, the proposed DPD-YOLO demonstrated superior detection performance for pineapples in situations where the background is complex and the targets are occluded. The mAP@0.5 reached 62.0%, representing an improvement of 6.6% over the original YOLOv8 algorithm, Precision increased by 2.7%, Recall improved by 13%, and F1-score rose by 10.3%.

Underwater object detection plays a significant role in fisheries resource assessment and ecological environment protection. However, traditional underwater object detection methods struggle to achieve accurate detection in complex underwater environments with limited computational resources. This paper proposes a lightweight underwater object detection network called LightFusionNet-YOLO (LFN-YOLO). First, we introduce the reparameterization technique RepGhost to reduce the number of parameters while enhancing training and inference efficiency. This approach effectively minimizes precision loss even with a lightweight backbone network. Then, we replaced the standard depthwise convolution in the feature extraction network with SPD-Conv, which includes an additional pooling layer to mitigate detail loss. This modification effectively enhances the detection performance for small objects. Furthermore, We employed the Generalized Feature Pyramid Network (GFPN) for feature fusion in the network's neck, enhancing the network's adaptability to features of varying scales. Finally, we design a new detection head, CLLAHead, which reduces computational costs and strengthens the robustness of the model through cross-layer local attention. At the same time, the DFL loss function is introduced to reduce regression and classification errors. Experiments conducted on public datasets, including URPC, Brackish, and TrashCan, showed that the mAP@0.5 reached 74.1%, 97.5%, and 66.2%, respectively, with parameter sizes and computational complexities of 2.7M and 7.2 GFLOPs, and the model size is only 5.9 Mb. Compared to mainstream vision models, our model demonstrates superior performance. Additionally, deployment on the NVIDIA Jetson AGX Orin edge computing device confirms its high real-time performance and suitability for underwater applications, further showcasing the exceptional capabilities of LFN-YOLO.

Surface sediment samples were collected from the surrounding waters of the two largest tourist islands in Sanya, China, to compare and evaluate the sources, distribution, and ecological risks of 21 organochlorine pesticides (OCPs). The total concentration of OCPs ranged from 1.35 to 5.0 ng/g. Among the OCPs, ΣDDTs accounted for the largest proportion, followed by ΣHCHs. The concentrations of HCHs and DDTs from the west side of West Island were significantly higher than those from the east side, and fine-grained sediments exhibited a stronger adsorption effect on OCPs. Source analysis indicated that the area experienced new inputs of HCH pollutants, while historical residues of HCHs remained high. Residual OCPs are still widely present in the environment, transported mainly by river runoff, with a smaller portion originating from atmospheric deposition and ship paints. Ecological risk assessment results showed that factors occasionally causing adverse biological effects include Heptachlor epoxide, 4,4'-DDE, ΣDDT, Dieldrin, Endrin, and γ-HCH, while other factors rarely caused negative biological effects. Potential ecological effect evaluations indicated that stations SY03, SY04, SY06, and SY09 were classified as having moderate ecological effect levels, while other stations were classified as having no ecological effects. Strengthened investigation, monitoring, and control of pollutant sources in ecologically impacted areas are necessary. This study fills a data gap for the region and provides an academic foundation for environmental protection and the sustainable development of tourism resources.

Considerable differences in molecular characteristics have been defined between non-smoker and smokers in patients with lung adenocarcinoma (LUAD), yet studies on open chromatin patterns associated with LUAD progression caused by smoking are still lacking. Here, we constructed a novel network based on correlations between each ATAC-seq peak from TCGA data using our previously developed algorithm. Subsequently, principal component analysis was performed on LUAD samples with retained peaks filtered by the correlation network, and pathway analysis was conducted to identify potential pathways involved. We identified a set of peaks that discriminated smokers in LUAD patients according to levels of exposure to tobacco quantified in pack-years. These peaks were also significantly associated with progression-free survival and overall survival of these patients. Further examination of the gene set related to those peaks revealed that the comprising genes, such as KRT19, B3GNT3, CLDN7 and CLDN3 are strongly associated with LUAD development. They are consistent with the important roles of the associated pathways in LUAD oncogenesis induced by smoking, including estrogen response, apical junction and glycolysis pathways. In summary, our study may provide valuable insights into exploring ATAC-seq peaks and understanding smoking-related LUAD carcinogenesis from a perspective of open chromatin changes.

With the deep integration of artificial intelligence (AI) and education, the traditional medical education model has changed greatly. In view of the current application of AI technology represented by ChatGPT, machine learning in clinical medicine, changes should also be actively sought in the teaching of medical practice course. The reform necessity and research status of medical practice course teaching mode in the context of AI were elaborated from the aspects of teaching methods and effects evaluation in this study. And the reform trend for teaching mode of clinical practice course in the context of AI era from three aspects were discussed, innovation in teaching concepts and methodology, accurate teaching evaluation, and AI thinking training of teachers. This study provides a feasible way for medical colleges to train high-level medical talents to meet the development needs of the times.

Abstract Medical imaging is indispensable for accurate diagnosis and effective treatment, with modalities like MRI and CT providing diverse yet complementary information. Traditional image fusion methods, while essential in consolidating information from multiple modalities, often suffer from poor image quality and loss of crucial details due to inadequate handling of semantic information and limited feature extraction capabilities. This paper introduces a novel medical image fusion technique leveraging unsupervised image segmentation to enhance the semantic understanding of the fusion process. The proposed method, named DUSMIF, employs a multi-branch, multi-scale deep learning architecture that integrates advanced attention mechanisms to refine the feature extraction and fusion processes. An innovative approach that utilizes unsupervised image segmentation to extract semantic information is introduced, which is then integrated into the fusion process. This not only enhances the semantic relevance of the fused images but also improves the overall fusion quality. The paper proposes a sophisticated network structure that extracts and fuses features at multiple scales and across multiple branches. This structure is designed to capture a comprehensive range of image details and contextual information, significantly improving the fusion outcomes. Multiple attention mechanisms are incorporated to selectively emphasize important features and integrate them effectively across different modalities and scales. This approach ensures that the fused images maintain high quality and detail fidelity. A joint loss function combining content loss, structural similarity loss, and semantic loss is formulated. This function not only guides the network in preserving image brightness and texture but also ensures that the fused image closely resembles the source images in both content and structure. The proposed method demonstrates superior performance over existing fusion techniques in objective assessments and subjective evaluations, confirming its effectiveness in enhancing the diagnostic utility of fused medical images.

To understand the pollution characteristics and potential sources of heavy metal(loid)s in the nearshore sediments of Qionghai, 93 surface sediment samples were collected from the region. The concentrations of 20 elements, including Fe, Mg, Ca, Ti, Mn, Ba, Cr, Sr, Ni, Cu, Zr, As, Hg, Se, Be, Co, Mo, Cd, Ga, and Pb, were measured. The extent of contamination and ecological risk posed by these heavy metals/metalloids were evaluated using the geo-accumulation index, potential ecological risk index, and Nemerow comprehensive risk index. Additionally, correlation analysis, principal component analysis (PCA), and positive matrix factorization (PMF) were employed to identify the potential sources of these elements in the sediments. The findings reveal the following: (1) The mean concentrations of Fe, Ca, Mg, Ti, Cu, Sr, Zr, Mo, Cd, Pb, Hg, As, and Se exceed the background values for shallow sea sediments in China. Notably, Ca, Ti, Sr, Zr, Mo, Hg, and As exhibit coefficients of variation greater than 51%, indicating significant spatial variability primarily driven by anthropogenic activities. (2) The ecological risk assessment identifies Sr, Hg, and As as the principal pollutants and key potential ecological risk factors in the study area, necessitating prioritization in subsequent monitoring efforts. (3) Correlation and source analysis suggest that As and Mn primarily originate from agricultural activities, Sr, Ca, and Mg from aquaculture, Zr, Ti, Mo, Se, Pb, Be, Co, Cu, Ga, Ni, Fe, and Cd from natural sources, and Hg, Ba, and Cr from transportation sources. Additionally, this study identified Sr, Hg, and As as the primary pollutants in the Qionghai nearshore area, with sources predominantly linked to agriculture, aquaculture, and traffic. Regular monitoring will help track the effectiveness of implemented control measures and provide data for ongoing risk assessments, ensuring the protection and sustainability of the marine environment.

The utilization of the one-stop strategy for treating atrial fibrillation (AF) has been on the rise, but the combination of vein of Marshall catheter ablation with left atrial appendage occlusion has been rarely reported. In this case, a 62-year-old man with persistent AF underwent a one-stop strategy that included ethanol infusion in the vein of Marshall, radiofrequency catheter ablation, and left atrial appendage occlusion. The procedures successfully achieved ablation in the vein of Marshall, pulmonary vein isolation, and left atrial appendage occlusion. The feasibility and safety of this comprehensive strategy were evaluated during a 12-month follow-up. The dynamic ECG indicated the presence of sinus rhythm, and a cerebral CT scan confirmed the absence of silent ischemia. This case highlights a novel approach to treating atrial fibrillation and demonstrates promising outcomes for patients undergoing this combined procedure.

Abstract Dynamic Tensor Inversion (DTI) is an emerging issue in recent research, prevalent in artificial intelligence development frameworks such as TensorFlow and PyTorch. Traditional numerical methods suffer significant lagging error when addressing this issue. To address this, Zeroing-type Neural Dynamics (ZND) and Gradient-type Neural Dynamics (GND) are employed to tackle the DTI. However, these two methods exhibit inherent limitations in the resolution process, i.e. high computational complexity and low solution accuracy, respectively. Motivated by this technology gap, this paper proposes an Adaptive Coefficient Gradient Neural Dynamics (ACGND) for dynamically solving the DTI with an efficient and precise manner. Through a series of simulation experiments and validations in engineering applications, the ACGND demonstrates advantages in resolving DTI. The ACGND enhances computational efficiency by circumventing matrix inversion, thereby reducing computational complexity. Moreover, its incorporation of adaptive coefficients and activation functions enables real-time adjustments of the computational solution, facilitating rapid convergence to theoretical solutions and adaptation to non-statinary scenarios. Code is available at https://github.com/Maia2333/ACGND-Code-Implementation .

Automated harvesting of “Sunshine Rose” grapes requires accurate detection and classification of grape clusters under challenging orchard conditions, such as occlusion and variable lighting, while ensuring that the model can be deployed on resource- and computation-constrained edge devices. This study addresses these challenges by proposing a lightweight YOLOv8-based model, incorporating DualConv and the novel C2f-GND module to enhance feature extraction and reduce computational complexity. Evaluated on the newly developed Shine-Muscat-Complex dataset of 4715 images, the proposed model achieved a 2.6% improvement in mean Average Precision (mAP) over YOLOv8n while reducing parameters by 36.8%, FLOPs by 34.1%, and inference time by 15%. Compared with the latest YOLOv11n, our model achieved a 3.3% improvement in mAP, with reductions of 26.4% in parameters, 14.3% in FLOPs, and 14.6% in inference time, demonstrating comprehensive enhancements. These results highlight the potential of our model for accurate and efficient deployment on resource-constrained edge devices, providing an algorithmic foundation for the automated harvesting of “Sunshine Rose” grapes.

In order to accurately predict the power consumption data of charging piles, assist related enterprises to accurately predict the benefits of charging piles and further optimize the relationship between households and transformers, this paper proposes an improved Gate Recurrent Unit (IGRU) prediction model based on spline interpolation. This method first extracts relevant data features based on Pearson correlation analysis, and then fuses feature data and performs spline interpolation input prediction network model for charging pile load pre-diction. Finally, based on the power consumption data of 174 charging piles in the jurisdiction of a provincial capital city in southwest China, the prediction accuracy of the model is verified. Through case analysis and experimental comparison, it shows that the improve gate recurrent unit (IGRU) prediction model designed in this paper is superior to the traditional deep learning prediction model in terms of stability, prediction accuracy and generalization ability.

With the establishment of smart grids, there is a growing demand for metering electric meters in urban areas. Given the diverse sizes and delicate nature of these meters, traditional scheduling solutions are unable to cater to the multifaceted requirements of urban environments, meters loading, and subsequent logistics scheduling. This study presents an intelligent scheduling model for electric meters in an urban context, taking into account various constraints such as urban traffic congestion, three-dimensional packing of metering devices, delivery time windows, and heterogeneous vehicles. To solve this, we design an improved whale optimization algorithm using a hybrid multi-phase heuristic approach (IWOA-HMOHA). Simulation results show that compared with the traditional meter logistics scheduling strategy, the IWOA-HMOHA algorithm reduces the objective function by 5.4%~ 26.1% compared with other similar algorithms. In addition, compared with the traditional first-in-last-out cargo packing method, the vehicle space utilization rate is improved by 12.62%. The proposed models and algorithms demonstrate excellent adaptability to a range of urban constraints, offering valuable insights and a robust framework essential for the development of logistics solutions in urban.

In optimal control, it is often necessary to solve Hamilton-Jacobi-Isaacs (HJI) partial differential equation, but it is not only difficult to solve, sometimes even impossible to solve. It is possible to avoid solving the HJI equation by using inverse optimal methods. We investigate inverse optimality of regulation design for Korteweg-de Vries-Burgers (KdVB) equation. Two kinds of boundary control laws are achieved to regulate the state of closed-loop system to the set point from any initial value. In order to regulate the convergent speed of the closed-loop system, one or two parameters are designed in the boundary control laws. We proved that boundary control laws are optimal for two meaningful functionals, respectively. The effectiveness of the proposed design has been shown through simulations, and the convergence speed of the closed-loop system accelerates with increase of adjustable parameters.

Super-resolution reconstruction technology is a crucial approach to enhance the quality of remote sensing optical images. Currently, the mainstream reconstruction methods leverage convolutional neural networks (CNNs). However, they overlook the global information of the images, thereby impacting the reconstruction effectiveness. Methods based on Transformer networks have demonstrated the capability to improve reconstruction quality, but the high model complexity renders them unsuitable for remote sensing devices. To enhance reconstruction performance while maintaining the model lightweight, a distillation Transform-CNN Network is proposed in this article. The strategy employs the Transformer network as a teacher network, guiding its long-range features into a compact CNN, achieving distillation across networks. Simultaneously, to rectify misinformation in the teacher network, prior information is introduced to ensure accurate information transfer. Concerning the student network, a novel upsampling approach is devised, utilizing inherent information in downsampled feature maps for padding, thereby avoiding the introduction of zero-information feature points in the traditional deconvolution process. Experimental evaluations conducted on multiple publicly available remote sensing image datasets demonstrate that the proposed method, while maintaining a smaller parameter count, achieves outstanding reconstruction quality for remote sensing images, surpassing existing approaches.

Accurate and timely information on urban vegetation (UV) can be used as an important indicator to estimate the health of cities. Due to the low cost of RGB cameras, true color imagery (TCI) has been widely used for high spatial resolution UV mapping. However, the current index-based and classifier-based UV mapping approaches face problems of the poor ability to accurately distinguish UV and the high reliance on massive annotated samples, respectively. To address this issue, an index-guided semantic segmentation (IGSS) framework is proposed in this paper. Firstly, a novel cross-scale vegetation index (CSVI) is calculated by the combination of TCI and Sentinel-2 images, and the index value can be used to provide an initial UV map. Secondly, reliable UV and non-UV samples are automatically generated for training the semantic segmentation model, and then the refined UV map can be produced. The experimental results show that the proposed CSVI outperformed the existing five RGB vegetation indices in highlighting UV cover and suppressing complex backgrounds, and the proposed IGSS workflow achieved satisfactory results with an OA of 87.72% ∼ 88.16% and an F1 score of 87.73% ∼ 88.37%, which is comparable with the fully-supervised method.

Abstract Perovskite-type lead-free piezoelectric ceramics allow access to illustrious piezoelectric coefficients (d 33) through intricate composition design and experimental modulation. Developing a swift and accurate technology for identifying (K, Na)NbO3 (KNN)-based ceramic compositions with high d 33 in exceedingly large “compositional” space will establish an innovative research paradigm surpassing the traditional empirical trial-and-error method. Herein, we demonstrate an interpretable machine learning (ML) framework for quick evaluation of KNN-based ceramics with high d 33 based on data from published literature. Specifically, a thorough feature construction was carried out from the global and local dimensions to establish tree regression models with d 33 as the target property. Subsequently, the feature-property mapping rules of KNN-based piezoelectric ceramics are further optimized through feature screening. To intuitively understand the correlation mechanisms between ML regression targets and features, the sure independence screening and sparsifying operator (SISSO) method was employed to extract the essential descriptors to explain d 33. A straightforward descriptor, $${\text{e}}^{({{NM}}_{\text{B}}-{{MV}}_{\text{B}})}\cdot {ST}/{(I{D}_{\text{A}})}^{2}$$ e ( NM B − MV B ) ⋅ ST / ( I D A ) 2 , consisting of only four easily accessible parameters, can accelerate the evaluation of a series of novel KNN-based ceramics with high d 33 while exhibiting strong theoretical interpretability. This work not only provides a tool for the rapid discovery of high piezoelectric performance in KNN-based ceramics but also offers a data-driven route for the design of property descriptors in perovskites.

Abstract Structural superlubricity, a state of nearly zero friction and no wear between two contact surfaces under relative sliding, holds immense potential for research and application prospects in micro-electro-mechanical systems devices, mechanical engineering, and energy resources. A critical step towards the practical application of structural superlubricity is the mass transfer and high throughput performance evaluation. Limited by the yield rate of material preparation, existing automated systems, such as roll printing or massive stamping, are inadequate for this task. In this paper, a machine learning-assisted system is proposed to realize fully automated selective transfer and tribological performance measurement for structural superlubricity materials. Specifically, the system has a judgment accuracy of over 98% for the selection of micro-scale graphite flakes with structural superlubricity properties and complete the 100 graphite flakes assembly array to form various pre-designed patterns within 100 mins, which is 15 times faster than manual operation. Besides, the system is capable of automatically measuring the tribological performance of over 100 selected flakes on Si3N4, delivering statistical results for new interface which is beyond the reach of traditional methods. With its high accuracy, efficiency, and robustness, this machine learning-assisted system promotes the fundamental research and practical application of structural superlubricity.

[Introduction] With the rapid development of AC/DC distribution networks and distributed generation technology, the role of DC microgrids in distribution networks is becoming increasingly important and will become an important component of future distribution networks. Due to the small coverage area and low line impedance of the DC microgrid, when an inter pole short circuit fault occurs, the fault current increases rapidly and has a large amplitude, which can reach more than 10 times the rated working current. This makes it difficult to set the protection of DC microgrids and requires high equipment selection, which restricts the rapid development of DC microgrids. [Method] In response to the above issues, taking the DC microgrid as the research object, starting from the working principle of inter pole faults in the DC microgrid, the fault characteristics on the DC side of the DC microgrid were analyzed. In response to the shortcomings of existing main current limiting methods, a voltage controllable fault current controller was proposed to achieve precise control of fault current. The simulation model of DC microgrid and fault current controller was built for simulation verification. [Result] The simulation results show that the fault current controller can significantly reduce the fault current and achieve precise control of the fault current, making the system controllable before and after the fault without locking the protection. During steady-state operation, the fault current controller can also assist the VSC (Voltage Source Converter) in further stabilizing the DC bus voltage. [Conclusion] To cooperate with the normal operation of the relay protection device and avoid VSC triggering overcurrent protection blocking, it is recommended to set the fault current control range between 1~2 pu.

Abstract As one of the most promising alternatives to graphite negative electrodes, silicon oxide (SiO x ) has been hindered by its fast capacity fading. Solid electrolyte interphase (SEI) aging on silicon SiO x has been recognized as the most critical yet least understood facet. Herein, leveraging 3D focused ion beam-scanning electron microscopy (FIB-SEM) tomographic imaging, we reveal an exceptionally characteristic SEI microstructure with an incompact inner region and a dense outer region, which overturns the prevailing belief that SEIs are homogeneous structure and reveals the SEI evolution process. Through combining nanoprobe and electron energy loss spectroscopy (EELS), it is also discovered that the electronic conductivity of thick SEI relies on the percolation network within composed of conductive agents (e.g., carbon black particles), which are embedded into the SEI upon its growth. Therefore, the free growth of SEI will gradually attenuate this electron percolation network, thereby causing capacity decay of SiO x . Based on these findings, a proof-of-concept strategy is adopted to mechanically restrict the SEI growth via applying a confining layer on top of the electrode. Through shedding light on the fundamental understanding of SEI aging for SiO x anodes, this work could potentially inspire viable improving strategies in the future.

BackgroundMechanical thrombectomy (MT) is one of the effective treatment methods for acute ischemic stroke (AIS), which requires a period of dual antiplatelet therapy (DAPT) after endovascular treatment. This study aimed to compare the efficacy and safety of 3-month DAPT and 1-month DAPT in AIS patients receiving MT through a retrospective study.MethodsAIS patients who received MT from May 2018 to March 2023 were grouped into a 1-month group (1-M group) and a 3-month group (3-M group) according to the duration of DAPT after MT. The primary outcome was the mRS score at 90 days. Secondary outcomes included a good prognosis (mRS score of 0–2) at 90 days post-surgery, 6-month mortality, recurrence of cerebral infarction, Barthel's index, Montreal Cognitive Assessment (MoCA) score, and incidence of symptomatic intracranial hemorrhage (sICH) during hospitalization.ResultA total of 147 patients with AIS were included in the final analysis, with 78 patients in the 1-M group and 69 patients in the 3-M group. The baseline and neurological characteristics were comparable between both groups. At 3-month follow-up, a total of 61 patients had an mRS of 0–2 at 90 days, with an average mRS of 3.3 ± 0.9 for all patients. There was no statistically significant difference in the mRS between the two groups of patients at 90 days (P > 0.05). There was no statistically significant difference in the mortality rate and incidence of sICH between the two groups of patients during the 6-month follow-up period (P > 0.05), but the recurrence rate of AIS in the 3-M group was lower than that in the 1-M group (P < 0.05). The improvement of Barthel index and MoCA in patients in the 3-M group was higher than those in the 1-M group at 6 months but not statistically different (P > 0.05).ConclusionFor AIS patients undergoing mechanical thrombectomy, compared to 1-month DAPT, 3-month DAPT can reduce the recurrence rate of IS during a 6-month follow-up period, without increasing the mortality and risk of cerebral hemorrhage.

A phase-locked loop (PLL) of a phase-shifting power supply for the power system is designed in this paper. Taking the industrial frequency current as the reference signal, the input and output characteristics of the phase-frequency detector with charge pump are analyzed. In addition, the transfer function of the PLL system is analyzed by combining the passive second-order low-pass filter and the voltage-controlled oscillator. As a result, the influence of parameters such as charge pump current, phase margin, and loop bandwidth on the performance of the PLL is investigated, and the optimal parameters of each section are obtained. The results show that the optimized PLL has a locking time of 0.75 s, a pull-out range of 63.16 Hz, and a locking range of 43.82 Hz, with high stability and noise immunity.

In this work, the frequency dependence of ferroelectric and electrocaloric properties in barium titanate-based ceramics was studied based on Maxwell relations. It is found that the maximum and remnant polarization will decrease while the coercive field increases a lot with rising frequency from 0.1 to 10[Formula: see text]Hz, indicating that polarization rotation and domain switching become difficult at high frequencies. The electrocaloric properties show the different frequency dependence at different phase structures. Isothermal entropy change ([Formula: see text]) and adiabatic temperature change ([Formula: see text]) are similar around/above Curie temperature ([Formula: see text], showing tiny frequency dependence. However, [Formula: see text] and [Formula: see text] display the obvious frequency dependence below [Formula: see text], especially in the orthorhombic–tetragonal phase-transition region with a stable ferroelectric phase, and this frequency dependence becomes more obvious under a low-electric field. It is also found that increasing the frequency can weaken the electric field dependence of electrocaloric strength. This work gives a general profile of frequency dependence for electrocaloric properties in ferroelectric ceramics.

The project of China initiative Accelerator Driven Subcritical (CiADS) system has been started to construct in southeast China's Guangdong province since 2019, which is expected to be checked and accepted in the year 2025. In order to make the students in University of Chinese Academy of Sciences (UCAS) better understand the main characteristic and the operation condition in the subcritical nuclear facility, the training platform for CiADS has been developed based on the Cave Automatic Virtual Environment (CAVE) in the Institute of Modern Physics Chinese Academy of Sciences (IMPCAS). The CAVE platform is a kind of non-head mounted virtual reality display system, which can provide the immersive experience and the alternative training platform to substitute the dangerous operation experiments with strong radioactivity. In this paper, the CAVE platform for the training scenarios in CiADS system has been presented with real-time simulation feature, where the required devices to generate the auditory and visual senses with the interactive mode have been detailed. Moreover, the three dimensional modeling database has been created for the different operation conditions, which can bring more freedom for the teachers to generate the appropriate training courses for the students. All the user-friendly features will offer a deep realistic impression to the students for the purpose of getting the required knowledge and experience without the large costs in the traditional experimental nuclear reactor.

As industrial technology continues to advance through integration, society's demand for electricity is rapidly increasing. To meet the requirements of refined grid management and address the elevated challenges arising from the increased electrical load, this paper delves into the investigation of distribution vehicle scheduling for the practical scenario of batch rotation of smart meters. Initially, based on the practical distribution task requirements of a provincial metrology verification center, a multi-level optimization model is constructed for the batch rotation and distribution vehicle scheduling of smart meters. The primary objective is to maximize the enhancement of smart meter distribution efficiency while minimizing the overall distribution cost. Moreover, this paper introduces a refined Grey Wolf Optimization algorithm (OLC-GWO) based on Opposition-based Learning, Levy flight strategy, and Cauchy mutation to solve the model. By generating an opposite population to improve the quality of initial feasible solutions and further harnessing the global search capabilities of Levy flight and Cauchy mutation operators, the algorithm's effectiveness is enhanced. The algorithm is subjected to testing using multiple benchmark functions and its performance is compared with variants of GWO, as well as several cutting-edge intelligent optimization algorithms including Particle Swarm Optimization (PSO), Harris Hawks Optimization (HHO), and Honey Bee Algorithm (HBA). The results indicate that OLC-GWO exhibits excellent performance in terms of convergence speed and optimization capability. Finally, the improved algorithm is subjected to simulation experiments by incorporating order data from the practical distribution operations of a provincial metrology verification center. The outcomes verify the efficiency of the proposed algorithm, reinforcing the practical significance of the established model in addressing the real-world challenge of batch rotation and distribution vehicle scheduling for smart meters.

AIGC is currently a hot field and a future trend in AI applications, and addressing the challenge of digitally reconstructing cultural heritage under the influence of AI technology is a pressing issue that requires immediate resolution. The article proposes an application framework for AIGC technology that is based on refining its meaning and designing a specific process for applying it to the digital reconstruction of cultural heritage. A high-definition camera is used to acquire relevant images of cultural heritage. The image features are extracted by the SIFT algorithm optimized by the PROSAC algorithm. The color features are acquired by combining the color histogram, color moment, and color correlation diagram. The 3D laser scanning technology is used to obtain the 3D point cloud data of the cultural heritage; the KD-tree improved ICP algorithm is introduced to improve the efficiency of point cloud alignment; the dense reconstruction of the 3D point cloud data of the cultural heritage is realized based on CMVS/PMVS; and the immersive 3D experience system of the cultural heritage is constructed by combining with platforms such as Unity3D. The average matching rate of the optimized SITF algorithm to the image features of cultural heritage is about 74.91%, and the maximum alignment time of the ICP algorithm to the cultural heritage point cloud data based on KD-tree is 9.241 s. The cultural heritage immersive 3D experience system has a satisfaction rate of 56.75%, and the density reconstructed model’s surface has an average deviation of only 0.34 mm from the real surface. The user satisfaction rating for the immersive 3D experience system for cultural heritage is 56.75%. Based on AIGC technology, it can revitalize cultural heritage and achieve digital reconstruction and inheritance innovation of cultural heritage.

It is essential for realizing the most suitable product buffer packaging design to quantify the vibration transmission characteristics of the product packaging system. The experiment system for the vibration transmission path of protective packaging is designed in this paper. The practical system is used to analyze the vibration transfer path of the product packaging system and identify the critical transfer path. The concepts of the cushions’ contribution rate and the cushions’ weighted contribution rate are introduced. The product cushioning based on the weighted equal contribution rate of the cushions is proposed. It has been verified by experiments that the system can accurately identify the transfer path with the weighted contribution rate of the cushions as a reference for the design of product buffer packaging, which improves the utilization rate of buffer packaging materials and reduces the cost of packaging materials. The weighted equal contribution rates of buffer pads 1, 2, 3, and 4 are 40%, 27%, 22%, and 11%, respectively. For the needs of experiment teaching, the teaching content based on the protective packaging transfer path testing system is designed, which provides a reference for the practical education of the packaging specialty.

Abstract This paper presents a monolithically integrated aptasensor composed of a piezoresistive microcantilever array and an on-chip signal processing circuit. Twelve microcantilevers, each of them embedded with a piezoresistor, form three sensors in a Wheatstone bridge configuration. The on-chip signal processing circuit consists of a multiplexer, a chopper instrumentation amplifier, a low-pass filter, a sigma-delta analog-to-digital converter, and a serial peripheral interface. Both the microcantilever array and the on-chip signal processing circuit were fabricated on the single-crystalline silicon device layer of a silicon-on-insulator (SOI) wafer with partially depleted (PD) CMOS technology followed by three micromachining processes. The integrated microcantilever sensor makes full use of the high gauge factor of single-crystalline silicon to achieve low parasitic, latch-up, and leakage current in the PD-SOI CMOS. A measured deflection sensitivity of 0.98 × 10− 6 nm−1 and an output voltage fluctuation of less than 1 μV were obtained for the integrated microcantilever. A maximum gain of 134.97 and an input offset current of only 0.623 nA were acquired for the on-chip signal processing circuit. By functionalizing the measurement microcantilevers with a biotin-avidin system method, human IgG, abrin, and staphylococcus enterotoxin B (SEB) were detected at a limit of detection (LOD) of 48 pg/mL. Moreover, multichannel detection of the three integrated microcantilever aptasensors was also verified by detecting SEB. All these experimental results indicate that the design and process of monolithically integrated microcantilevers can meet the requirements of high-sensitivity detection of biomolecules.

Toll-like receptors (TLRs) play a key role in the innate immune system. Here, Kornilov et al. resolve the 3D structures of the membrane-associated parts of four TLRs to reveal properties of the juxta-membrane domain.

Abstract The accurate detection of leukocytes is the basis for the diagnosis of blood system diseases. However, diagnosing leukocyte disorders by doctors is time-consuming and requires extensive experience. Automated detection methods with high accuracy can improve detection efficiency and provide recommendations to inexperienced doctors. Current methods and instruments either fail to automate the identification process fully or have low performance and need suitable leukocyte data sets for further study. To improve the current status, we need to develop more intelligent strategies. This paper investigates fulfilling high-performance automatic detection for leukocytes using a deep learning-based method. We established a new dataset more suitable for leukocyte detection, containing 6273 images (8595 leukocytes) and considering nine common clinical interference factors. Based on the dataset, the performance evaluation of six mainstream detection models is carried out, and a more robust ensemble model is proposed. The mean of average precision (mAP) @IoU = 0.50:0.95 and mean of average recall (mAR)@IoU = 0.50:0.95 of the ensemble model on the test set are 0.853 and 0.922, respectively. The detection performance of poor-quality images is robust. For the first time, it is found that the ensemble model yields an accuracy of 98.84% for detecting incomplete leukocytes. In addition, we also compared the test results of different models and found multiple identical false detections of the models, then provided correct suggestions for the clinic.

The core power control is an important issue for the study of dynamic characteristics in China initiative accelerator driven subcritical system (CiADS), which has direct impact on the control strategy and safety analysis process. The CiADS is an experimental facility that is only controlled by the proton beam intensity without considering the control rods in the current engineering design stage. In order to get the optimized operation scheme with the stable and reliable features, the variation of beam intensity using the continuous and periodic control approaches has been adopted, and the change of collimator and the adjusting of duty ratio have been proposed in the power control process. Considering the neutronics and the thermal-hydraulics characteristics in CiADS, the physical model for the core power control has been established by means of the point reactor kinetics method and the lumped parameter method. Moreover, the multi-inputs single-output (MISO) logical structure for the power control process has been constructed using proportional integral derivative (PID) controller, and the meta-heuristic algorithm has been employed to obtain the global optimized parameters for the stable running mode without producing large perturbations. Finally, the verification and validation of the control method have been tested based on the reference scenarios in considering the disturbances of spallation neutron source and inlet temperature respectively, where all the numerical results reveal that the optimization method has satisfactory performance in the CiADS core power control scenarios.

Most of the research on disease recognition in chest X-rays is limited to segmentation and classification, but the problem of inaccurate recognition in edges and small parts makes doctors spend more time making judgments. In this paper, we propose a lesion detection method based on a scalable attention residual CNN (SAR-CNN), which uses target detection to identify and locate diseases in chest X-rays and greatly improves work efficiency. We designed a multi-convolution feature fusion block (MFFB), tree-structured aggregation module (TSAM), and scalable channel and spatial attention (SCSA), which can effectively alleviate the difficulties in chest X-ray recognition caused by single resolution, weak communication of features of different layers, and lack of attention fusion, respectively. These three modules are embeddable and can be easily combined with other networks. Through a large number of experiments on the largest public lung chest radiograph detection dataset, VinDr-CXR, the mean average precision (mAP) of the proposed method was improved from 12.83% to 15.75% in the case of the PASCAL VOC 2010 standard, with IoU > 0.4, which exceeds the existing mainstream deep learning model. In addition, the proposed model has a lower complexity and faster reasoning speed, which is conducive to the implementation of computer-aided systems and provides referential solutions for relevant communities.

Abstract The noise and redundant information are the main reasons for the performance bottleneck of medical image segmentation algorithms based on the deep learning. To this end, we propose a deep network embedded with rough fuzzy discretization (RFDDN) for OCT fundus image segmentation. Firstly, we establish the information decision table of OCT fundus image segmentation, and regard each category of segmentation region as a fuzzy set. Then, we use the fuzzy c-means clustering to get the membership degrees of pixels to each segmentation region. According to membership functions and the equivalence relation generated by the brightness attribute, we design the individual fitness function based on the rough fuzzy set, and use a genetic algorithm to search for the best breakpoints to discretize the features of OCT fundus images. Finally, we take the feature discretization based on the rough fuzzy set as the pre-module of the deep neural network, and introduce the deep supervised attention mechanism to obtain the important multi-scale information. We compare RFDDN with U-Net, ReLayNet, CE-Net, MultiResUNet, and ISCLNet on the two groups of 3D retinal OCT data. RFDDN is superior to the other five methods on all evaluation indicators. The results obtained by ISCLNet are the second only inferior to those obtained by RFDDN. DSC, sensitivity, and specificity of RFDDN are evenly 3.3%, 2.6%, and 7.1% higher than those of ISCLNet, respectively. HD95 and ASD of RFDDN are evenly 6.6% and 19.7% lower than those of ISCLNet, respectively. The experimental results show that our method can effectively eliminate the noise and redundant information in Oct fundus images, and greatly improve the accuracy of OCT fundus image segmentation while taking into account the interpretability and computational efficiency.

Remarkable effectiveness has been demonstrated by deep neural networks in the despeckling task for synthetic aperture radar (SAR) images. However, blurring and loss of fine details can result from many despeckling models due to upsampling and mean-square-error (MSE) loss. Additionally, existing degradation models and prior information are ignored by existing despeckling models, which directly learn the mapping from degraded to clear images. To address these issues, an optimization algorithm for the SAR despeckling task based on the integral-Newton method is proposed in this article. Then, a prior-driven despeckling network is proposed, which can automatically capture the implicit priors in SAR images to replace traditional manually made priors. Furthermore, to make the network focus more on learning the structural prior information of images, a structure-preserving loss function based on the MSE and the Canny edge detection operator is designed, which improves the detail of the network retention ability and speeds up convergence. Outstanding results on both simulated datasets and real SAR images are achieved by the proposed method, as shown by a large number of experimental results. Moreover, significant advantages of the proposed method both visually and quantitatively are revealed by comparison with classical and state-of-the-art despeckling algorithms.

Abstract Zeroing neural networks (ZNN) have shown their state-of-the-art performance on dynamic problems. However, ZNNs are vulnerable to perturbations, which causes reliability concerns in these models owing to the potentially severe consequences. Although it has been reported that some models possess enhanced robustness but cost worse convergence speed. In order to address these problems, a robust neural dynamic with an adaptive coefficient (RNDAC) model is proposed, aided by the novel adaptive activation function and robust evolution formula to boost convergence speed and preserve robustness accuracy. In order to validate and analyze the performance of the RNDAC model, it is applied to solve the dynamic matrix square root (DMSR) problem. Related experiment results show that the RNDAC model reliably solves the DMSR question perturbed by various noises. Using the RNDAC model, we are able to reduce the residual error from 10 $$^1$$ 1 to 10 $$^{-4}$$ - 4 with noise perturbed and reached a satisfying and competitive convergence speed, which converges within 3 s.

Abstract In recent years, cannabidiol (CBD), a non‐psychotropic cannabinoid, has garnered substantial interest in drug development due to its broad pharmacological activity and multi‐target effects. Diabetes is a chronic metabolic disease that can damage multiple organs in the body, leading to the development of complications such as abnormal kidney function, vision loss, neuropathy, and cardiovascular disease. CBD has demonstrated significant therapeutic potential in treating diabetes mellitus and its complications owing to its various pharmacological effects. This work summarizes the role of CBD in diabetes and its impact on complications such as cardiovascular dysfunction, nephropathy, retinopathy, and neuropathy. Strategies for discovering molecular targets for CBD in the treatment of diabetes and its complications are also proposed. Moreover, ways to optimize the structure of CBD based on known targets to generate new CBD analogues are explored.

The online calibration method for current transformers is an important research direction in the field of smart grids. This article constructs a transformer error prediction model based on spatiotemporal integration. This model draws inspiration from the structure of forgetting gates in gated loop units and combines it with a graph convolutional network (GCN) that is good at capturing the spatial relationships within the graph attention network to construct an adaptive GCN. The spatial module formed by this adaptive GCN is used to model the spatial relationships in the circuit network, and the attention mechanism and gated time convolutional network are combined to form a time module to learn the temporal relationships in the circuit network. The layer that combines the time and space modules is used, which consists of a gating mechanism for spatiotemporal fusion, and a transformer error prediction model based on a spatiotemporal correlation analysis is constructed. Finally, it is verified on a real power grid operation dataset, and compared with the existing prediction methods to analyze its performance.

Abstract Background Systemic sclerosis (SSc) is an incurable autoimmune disease characterized by progressive skin fibrosis and organ failure. Tenosynovitis is a common musculoskeletal manifestation, but tendon rupture has seldom reported in SSc. Case presentation We present a rare case of a 49-year-old female with SSc who has suffered from bilateral tendon rupture of the fourth and fifth digits with positive antinuclear antibody (ANA) and anti-centromere B antibody, but negative rheumatoid factor in serum. In the extensor tendons of the patient’s hands, inflammation, edema, hypertrophy and tendon interruption were detected with ultrasound and magnetic resonance imaging(MRI). Tendon transfer repair surgery was performed and 10 mg/week methotrexate was then used in this patient. Her hand function was improved well with methotrexate and rehabilitation treatment postoperatively. Conclusions Early detection of tenosynovitis is necessary to prevent tendon rupture in SSc patients. Ultrasound and Magnetic Resonance Imaging appear to be useful examinations for evaluating tendon pathology for early detection.

Bayesian Hierarchical models has been widely used in modern statistical application. To deal with the data having complex structures, we propose a generalized hierarchical normal linear (GHNL) model which accommodates arbitrarily many levels, usual design matrices and ‘vanilla’ covariance matrices. Objective hyperpriors can be employed for the GHNL model to express ignorance or match frequentist properties, yet the common objective Bayesian approaches are infeasible or fraught with danger in hierarchical modelling. To tackle this issue, [Berger, J., Sun, D., & Song, C. (2020b). An objective prior for hyperparameters in normal hierarchical models. Journal of Multivariate Analysis, 178, 104606. https://doi.org/10.1016/j.jmva.2020.104606] proposed a particular objective prior and investigated its properties comprehensively. Posterior propriety is important for the choice of priors to guarantee the convergence of MCMC samplers. James Berger conjectured that the resulting posterior is proper for a hierarchical normal model with arbitrarily many levels, a rigorous proof of which was not given, however. In this paper, we complete this story and provide an user-friendly guidance. One main contribution of this paper is to propose a new technique for deriving an elaborate upper bound on the integrated likelihood, but also one unified approach to checking the posterior propriety for linear models. An efficient Gibbs sampling method is also introduced and outperforms other sampling approaches considerably.

Background: Whole-genome bisulfite sequencing (WGBS) technology can provide comprehensive DNA methylation at a single-base resolution on a genome-wide scale, and is considered to be the gold standard for the detection of 5-methylcytosine (5 mC). However, the International Human Epigenome Consortium propose a full DNA methylome should have at least 30 fold redundant coverage of the reference genome from a single biological replicate. Therefore, it remains cost prohibitive for large-scale studies. To find a solution, the DNBSEQ-Tx sequencing was developed that can generate up to 6 Tb data in a single run for projects involving large-scale sequencing. Results: In this study, we provided two WGBS library construction methods DNB_PREBSseq and DNB_SPLATseq optimized for the DNBSEQ-Tx sequencer, and demonstrated the performance of these two methods on the DNBSEQ-Tx platform, using the DNA extracted from four different cell lines. We also compared the sequencing data from these two WGBS library construction methods with HeLa cell line data from ENCODE sequenced on Illumina HiSeq X Ten and WGBS data of two other cell lines sequenced on HiSeq2500. Various quality control (QC) analyses such as the base quality scores, methylation-bias (m-bias), and conversion efficiency indicated that the data sequenced on the DNBSEQ-Tx platform met the WGBS-required quality controls. Meanwhile, our data closely resembled the coverage shown by the data generated by the Illumina platform. Conclusions: Our study showed that with our optimized methods, DNBSEQ-Tx could generate high-quality WGBS data with relatively good stability for large-scale WGBS sequencing applications. Thus, we conclude that DNBSEQ-Tx can be used for a wide range of WGBS research.

Abstract Systemic sclerosis (SSc) is an autoimmune rheumatic disease with high mortality, which is featured by inflammation, vascular damage, and aggressive fibrosis. To date, the pathogenesis of SSc remains unclear and effective treatments are still under research. Active NLRP3 recruits downstream proteins such as ASC and caspase-1 and assembles into inflammasome, resulting in excretion of inflammatory cytokines including IL-1β and IL-18, as well as in pyroptosis mediated by gasdermin D. Various studies demonstrated that NLRP3 inflammasome might be involved in the mechanism of tenosynovitis, arthritis, fibrosis, and vascular damage. The pathophysiological changes might be due to the activation of proinflammatory Th2 cells, profibrotic M2 macrophages, B cells, fibroblasts, and endothelial cells. Here, we review the studies focused on NLRP3 inflammasome activation, its association with innate and adaptive immune cells, endothelium injury, and differentiation of fibroblasts in SSc. Furthermore, we summarize the prospect of therapy targeting NLRP3 pathway.

Structure determination of covalent organic frameworks (COFs) is the key to pushing the development of COF-based materials further but precise determination of the structure of COFs is challenging. Here, the authors develop a universal ab initio structure determination method for polycrystalline 3D COFs using cryo-cRED by combining hierarchical cluster analysis with cryo-EM technique and demonstrate COF structures with atomic preciscion and up to 0.79-angstrom resolution.

The widespread use of pharmaceuticals and personal care products (PPCPs) in many fields has brought convenience to human lives but has also caused unavoidable environmental pollution issues. In particular, the resistance gene problem resulting from accumulating antibiotics that cannot be fully absorbed by biological individuals has been a concern; thus, it is urgent to find efficient technologies to boost the degradation efficiency of antibiotics in the environment. Here, an ε-MnO2 catalyst was prepared by a novel droplet-interface-drying method and utilized as a Fenton-like catalyst for efficiently degrading ciprofloxacin (CIP). The ε-MnO2 shell was formed preferentially at the gas–liquid interface and then continued to decompose into ε-MnO2 with abundant O vacancies in the air-insulated microcavity. The XPS result confirms that this particular preparation method can regulate the content of O vacancies in the material. Compared with ε-MnO2 samples obtained by the direct drying method (ε-MnO2-B), the catalytic performance of ε-MnO2 prepared by the droplet-interface-drying method (ε-MnO2-P) is significantly improved. By activating peroxymonosulfate (PMS) with the ε-MnO2-P catalyst, the CIP degradation efficiency can reach 84.1%. The detection and analysis of reactive oxygen species (ROS) in the ε-MnO2-P/PMS oxidation system confirms that ·OH, SO4·− and 1O2 are the main ROS for CIP degradation. This study highlights the creation of miniature hypoxic space to regulate the content of O vacancies in ε-MnO2, providing a new idea for the synthesis of other O-vacancy-rich materials.

For the unmet need to choose the suitable doublet detection method, an ensemble machine learning algorithm called Chord was developed, which integrates multiple methods and achieves higher accuracy and stability on different scRNA-seq datasets.

External stimuli induced effective regulations of luminescent material are of significant interest in the development of smart optical devices. Here, by simply doping with Er3+ in the 0.94Bi0.5Na0.5TiO3-0.06BaTiO3 (BNTBT) ferroelectric host, using the bendable mica substrate, and exerting mechanical strain (bending) or light illumination (via photochromic reaction), the all-inorganic, highly-transparent and flexible Er-doped BNTBT/Mica luminescent-ferroelectric thin films were designed and fabricated, displaying strain-induced dramatically elevation of up-conversion photoluminescence (PL) intensity, suppression of PL concentration quenching, outstanding endurance and durability, convenient illumination-mediated PL quenching. And the strain-induced structural changes and local lattice distortions of the thin films were further explored through theoretical calculations and Raman measurement. Our results can supply the guidance of designing other luminescent-ferroelectric materials with controlled PL properties via easy mechanical/photo stimuli for expanding the application of multifunctional wearable memory devices.

Abstract It is a long‐standing issue that the sluggish polysulfide conversion and adverse shuttling effects impede the development of lithium‐sulfur (Li‐S) batteries with high energy density and cycling stability, which necessitate the exploration of new electrocatalysts to facilitate the practical applications of Li‐S batteries. Herein, a single‐phase high‐entropy stabilized oxide (Ni0.2Co0.2Cu0.2Mg0.2Zn0.2)O (HEO850) is successfully prepared through a novel low‐temperature annealing strategy from a self‐sacrificing metal–organic frameworks (MOFs) template and then integrated into the sulfur host, where it functions as both the catalytic converter and chemical inhibitor towards the shuttle species. Furthermore, the synergistic contribution of randomly dispersed metal elements and the exposure of affluent active sites enable the chemical encapsulation of soluble polysulfides and accelerate conversion kinetics. The HEO850/S/KB cathode (KB: ketjen black; sulfur content: 70 wt.%) delivers a substantially higher initial specific discharge capacity of ~1244 mAh g−1 in comparison to MEO/S/KB (MEO: medium entropy oxide; ~980 mAh g−1), LEO/S/KB (LEO: low entropy oxide; ~908 mAh g−1), and routine S/KB cathodes (~966 mAh g−1), which is well retained at ~784 mAh g−1 after 800 cycles at 0.5 C with a low capacity decay rate of ~0.043% per cycle. Moreover, when the HEO850/S/KB cathode is processed with a high areal sulfur loading (~4.4 mg cm−2), the resulting Li‐S battery also performs well, with a high initial specific capacity of ~1044 mAh g−1 at 0.1 C and 85% capacity retention after 100 cycles. This study highlights the potential application of HEOs in enhancing the performance of Li‐S batteries and provides a novel strategy in synthesizing the HEOs at a relatively low annealing temperature for various energy conversion and storage applications.

Metabolic-associated fatty liver disease (MAFLD) is a chronic liver disease characterized by fatty infiltration of the liver. In recent years, the MAFLD incidence rate has risen and emerged as a serious public health concern. MAFLD typically progresses from the initial hepatocyte steatosis to steatohepatitis and then gradually advances to liver fibrosis, which may ultimately lead to cirrhosis and carcinogenesis. However, the potential evolutionary mechanisms still need to be clarified. Recent studies have shown that nucleotide methylation, which was directly associated with MAFLD’s inflammatory grading, lipid synthesis, and oxidative stress, plays a crucial role in the occurrence and progression of MAFLD. In this review, we highlight the regulatory function and associated mechanisms of nucleotide methylation modification in the progress of MAFLD, with a particular emphasis on its regulatory role in the inflammation of MAFLD, including the regulation of inflammation-related immune and metabolic microenvironment. Additionally, we summarize the potential value of nucleotide methylation in the diagnosis and treatment of MAFLD, intending to provide references for the future investigation of MAFLD.

We investigate stabilization of a class of cascaded systems of nonlinear ordinary differential equation (ODE)/wave partial differential equation (PDE) with time-varying propagation speed based on a two-step PDE backstepping transformation. A time-varying propagation velocity of wave PDE leads to two difficulties. One is how to prove the well-posedness and uniqueness of the time-varying kernel PDEs in the first-step backstepping transformation, the other is how to construct a backstepping transform to map the original system into a suitable target system during the second-step transformation. We prove that there exists a unique continuous $2 \times 2$ matrix-valued solution to the time-varying kernel PDEs, and design a predictor control for the original cascaded system. An example is provided to illustrate the feasibility of the proposed design.

Solving time-varying linear equation and inequality (TVLEI) problem has attracted extensive attention in numerous scientific and engineered fields. In this article, it is basically considered that the commonly used dynamics neural network in the virtual environment is inevitably interfered with by the variable measurement noises while dealing with the TVLEI problem. An adaptive enhanced and noise-suppressing zeroing neural network (AENSZNN) model is proposed as an improved algorithm for solving the TVLEI problem. An adaptive scale factor based on the residual error norm is designed to make the proposed AENSZNN model converge to the theoretical solution faster. Furthermore, the momentum enhancement terms added to the model enables the AENSZNN model to effectively solve the TVLEI problem in real-time under the obstruction of different measurement noises. Besides, theoretical results and numerical experiments indicate that the AENSZNN model has advantages in convergence accuracy and robustness to noises compared with the existing algorithms. Note that, the proposed AENSZNN model is successfully exploited for the estimation of mobile object localization.

We investigate output feedback stabilization for a class of high-order nonlinear systems whose output function and nonlinear terms are unknown. First, a smooth state feedback control law is designed by adding a power integrator technique. Next, we design a high-order observer to estimate the unmeasurable state, and allocate gains of the observer one by one in an iterative way. Finally, a dynamic output compensator is achieved such that the closed-loop system converges to the equilibrium point quick. Two examples are provided to demonstrate the effectiveness of the proposed method.