Search

This paper focuses on how regulatory bodies respond to artificial intelligence (AI)-enabled medical devices. To achieve this, we present a comparative overview of the United States (USA), European Union (EU), and China. Our search in the governmental database identified 59 AI medical devices approved in China as of July 2023. In comparison to the rules-based regulatory approach in China, the approaches in the USA and EU are more standards-oriented. [ABSTRACT FROM AUTHOR]

Purpose: Studying perivascular spaces (PVSs) is important for understanding the pathogenesis and pathological changes of neurological disorders. Although some methods for automated segmentation of PVSs have been proposed, most of them were based on 7T MR images that were majorly acquired in healthy young people. Notably, 7T MR imaging is rarely used in clinical practice. Herein, we propose a deep-learning-based method that enables automatic segmentation of PVSs on T2-weighted 3T MR images. Method: Twenty patients with Parkinson’s disease (age range, 42–79 years) participated in this study. Specifically, we introduced a multi-scale supervised dense nested attention network designed to segment the PVSs. This model fosters progressive interactions between high-level and low-level features. Simultaneously, it utilizes multi-scale foreground content for deep supervision, aiding in refining segmentation results at various levels. Result: Our method achieved the best segmentation results compared with the four other deep-learning-based methods, achieving a dice similarity coefficient (DSC) of 0.702. The results of the visual count of the PVSs in our model correlated extremely well with the expert scoring results on the T2-weighted images (basal ganglia: rs = 0.845, P < 0.001; rs = 0.868, P < 0.001; centrum semiovale: rs = 0.845, P < 0.001; rs = 0.823, P < 0.001 for raters 1 and 2, respectively). Experimental results show that the proposed method performs well in the segmentation of PVSs. Conclusion: The proposed method can accurately segment PVSs; it will facilitate practical clinical applications and is expected to replace the method of visual counting directly on T1-weighted images or T2-weighted images. [ABSTRACT FROM AUTHOR]

This study delves into Human–Computer Intelligent Interaction (HCII), a burgeoning interdisciplinary field that builds upon traditional Human–Computer Interaction (HCI) by integrating advanced technologies like Natural Language Processing (NLP) and Machine Learning (ML). In this paper, we scrutinize 5,781 HCII papers published between 2000 and 2023, narrowing our focus to 803 most relevant articles to construct co-citation and interdisciplinary networks based on the CiteSpace Software. Our findings reveal that the publications of the United States and China are relatively high with 558 and 616 publications respectively. Furthermore, we found that machine learning and deep learning have emerged as the prevalent methodologies in HCII, which currently emphasizes multimodal emotion recognition, facial expression recognition, and NLP. We predict that HCII will be integrated into advanced applications such as neural-based interactive games and multi-sensory environments. In sum, our analysis underscores HCII's role in advancing artificial intelligence, facilitating more intuitive and efficient human–computer interactions, and its prospective societal impact. We hope that our review and analysis may guide the efforts of researchers aiming to contribute to HCII and develop more powerful and intelligent methods, tools, and applications. [ABSTRACT FROM AUTHOR]

Accurate identification of tree species is the foundation of forest resource surveys and an important research field in forestry remote sensing. The introduction of the PointNet++ deep learning network to point cloud processing provides a new approach for tree species identification. This network can directly compute and train unordered point clouds, greatly reducing the manual selection and extraction process of feature data. However, how to establish a high-quality single tree point cloud dataset is a key issue to be solved. In this study, a 2.5-ha temperate coniferous broad-leaved mixed forest located in Mao'er Mountain Experimental Forest Farm in Heilongjiang Province, China, was investigated. The unmanned aerial vehicle (UAV) RGB image and LiDAR synchronous observation system were used to obtain the true-colour point clouds of the forest. Combining upsampling and downsampling methods, a dataset containing coordinates, normal vectors, RGB, and intensity information was constructed from the original point cloud. Comparative experiments were designed based on resampling algorithms, number of sampling points, feature information, and time cost to find the optimal feature information and processing methods for tree species classification. The results showed that the accuracy of tree species classification by using PointNet++ was greatly improved after adding RGB and point cloud intensity information. The classification accuracy was about 5% higher than that using only coordinate data sets. In addition, the combination of PU-Net downsampling network based on point cloud completion and geometric upsampling method achieved the highest classification accuracy (OA = 0.944) when the number of single trees point cloud was 3072. The algorithm also took a relatively shorter running time. This study demonstrated that the introduction of multi-feature information and the optimization of resampling method can provide new solutions for tree species classification based on the PointNet++ deep learning network. [ABSTRACT FROM AUTHOR]

This study addresses the limitations of existing indirect bridge damage identification methods that are based on the vehicle–bridge coupled vibration theory of highway bridges. To overcome these shortcomings, we propose an extended approach that incorporates various types of deep-learning models with vehicle–bridge coupled vibration responses. The proposed method is demonstrated using a three-span continuous beam bridge as a case study. First, a vehicle and bridge analysis model is established, and bridge damage is simulated using unit stiffness reduction, considering different damage scenarios. Next, to account for road roughness randomness, vehicle–bridge coupling vibration analysis is performed under various road roughness conditions, yielding the vertical acceleration vibration signal of the vehicle. Subsequently, we employ an end-to-end damage recognition method, utilizing the vehicle acceleration response as the network input, to construct two types of deep-learning models: one-dimensional convolutional neural network (1D-CNN) and convolutional long short-term memory neural network (CNN-LSTM). The recognition performance of both models is compared and analyzed. Taking Zhengzhou Taohuayu Self-Anchored Suspension Bridge in China as an example, this study delves into the capability of bridge damage identification using deep learning. The results demonstrate that the one-dimensional convolutional neural network achieves excellent recognition performance in terms of both damage location and severity. [ABSTRACT FROM AUTHOR]

Background: Investigating emotion sequence patterns in the posts of discussion forums in massive open online courses (MOOCs) holds a vital role in shaping online interactions and impacting learning achievement. While the majority of research focuses on the relationship between emotions and interactions in MOOC forum discussions, research on identifying the crucial difference in emotion sequence patterns among different interaction groups remains in its infancy. Objectives: This research utilizes deep learning and sequential pattern mining to investigate whether there are differences in emotion sequence patterns across different groups of learners who exhibit various types of interactions in online discussion forums. Methods: Data from a comprehensive array of sources, including log files, discussion texts and scores from 498 learners in online discussion forums, were collected for this study. The agglomerative hierarchical algorithm is used to classify learners into groups with different levels of interactions. Additionally, we implement and evaluate multiple deep learning models for detecting different emotions from online discussions. Relevant emotion sequence patterns were identified using sequence pattern analysis and the identified emotion sequence patterns were compared across different groups with different levels of interactions. Results and Conclusions: Using an agglomerative hierarchical algorithm, we classified learners into three distinct groups characterized by different levels of interactions: high, average and low level. Leveraging the bi‐directional long short‐term memory model for emotion detection yielded the highest predictive performance, with an impressive F‐measure of 94.01%, a recall rate of 93.83% and an accuracy score of 95.01%. The results also revealed that learners in the low‐level interaction group experienced more emotion transition from boredom to frustration than the other two groups. Therefore, the aggregation of students into groups and the utilization of their MOOC log data offer educators the capability to provide adaptive emotional feedback, customize assessments and offer more personalized attention as needed. Lay Description: What is currently known about the subject matter: Emotions are dynamic over time when learners experience cognitive disequilibrium/equilibrium.Online interactions are critical components, which influence learners' emotional state, cognitive processes and learning achievement.It is not clear what are differences in emotion sequence patterns across the groups with different interaction types. What the paper adds: An agglomerative hierarchical algorithm was implemented to cluster learners into three groups by analysing behavioural data.We explore possibilities for automated classification of emotions using deep learning approaches.Learners in the low‐level interaction group experienced more emotion transition from boredom to frustration. Implications for practitioners: A considerable amount of effort should be expended to identify and respond to learners who experience boredom and frustration emotions.Designing interventions or scaffolding to facilitate learners' interaction and promote favourable emotions.Educators could provide more personalized support based on learners' online interaction cluster. [ABSTRACT FROM AUTHOR]

Background: Peer assessment has played an important role in large‐scale online learning, as it helps promote the effectiveness of learners' online learning. However, with the emergence of numerical grades and textual feedback generated by peers, it is necessary to detect the reliability of the large amount of peer assessment data, and then develop an effective automated grading model to analyse the data and predict learners' learning results. Objectives: The present study aimed to propose an automated grading model with reliability detection. Methods: A total of 109,327 instances of peer assessment from a large‐scale teacher online learning program were tested in the experiments. The reliability detection approach included three steps: recurrent convolutional neural networks (RCNN) was used to detect grade consistency, bidirectional encoder representations from transformers (BERT) was used to detect text originality, and long short‐term memory (LSTM) was used to detect grade‐text consistency. Furthermore, the automated grading was designed with the BERT‐RCNN model. Results and Conclusions: The effectiveness of the automated grading model with reliability detection was shown. For reliability detection, RCNN performed best in detecting grade consistency with an accuracy rate of 0.889, BERT performed best in detecting text originality with an improvement of 4.47% compared to the benchmark model, and LSTM performed best with an accuracy rate of 0.883. Moreover, the automated grading model with reliability detection achieved good performance, with an accuracy rate of 0.89. Compared to the absence of reliability detection, it increased by 12.1%. Implications: The results strongly suggest that the automated grading model with reliability detection for large‐scale peer assessment is effective, with the following implications: (1) The introduction of reliability detection is necessary to help filter out low reliability data in peer assessment, thus promoting effective automated grading results. (2) This solution could assist assessors in adjusting the exclusion threshold of peer assessment reliability, providing a controllable automated grading tool to reducing manual workload with high quality. (3) This solution could shift educational institutions from labour‐intensive grading procedures to a more efficient educational assessment pattern, allowing for more investment in supporting instructors and learners to improve the quality of peer feedback. Lay Description: What is already known about this topic: Peer assessment has played an important role in large‐scale online learning, as it helps promote the effectiveness of learners' online learning.Issues such as disagreement between peer assessors, rough assessment, and plagiarism in large‐scale online learning can decrease peer assessment reliabilityIncorporating extensive data into a training model may result in grading uncertainties. What this paper adds: Detecting the peer assessment reliability before grading is essential in the context of large‐scale online learning.This study aimed to propose and validate an automated grading model with reliability detection for the large‐scale online peer assessment, which will help improve the effectiveness of automated grading, combining the advantages of computer technology and human expertise. Implications for practice and/or policy: The introduction of reliability detection is necessary to help filter out low reliability data in peer assessment, thus promoting effective automated grading results.This solution could assist assessors in adjusting the exclusion threshold of peer assessment reliability, providing a controllable automated grading tool to reducing manual workload with high quality. [ABSTRACT FROM AUTHOR]

This article elucidates the concept of large model technology, summarizes the research status of large model technology both domestically and internationally, provides an overview of the application status of large models in vertical industries, outlines the challenges and issues confronted in applying large models in the oil and gas sector, and offers prospects for the application of large models in the oil and gas industry. The existing large models can be briefly divided into three categories: large language models, visual large models, and multimodal large models. The application of large models in the oil and gas industry is still in its infancy. Based on open-source large language models, some oil and gas enterprises have released large language model products using methods like fine-tuning and retrieval augmented generation. Scholars have attempted to develop scenario-specific models for oil and gas operations by using visual/multimodal foundation models. A few researchers have constructed pre-trained foundation models for seismic data processing and interpretation, as well as core analysis. The application of large models in the oil and gas industry faces challenges such as current data quantity and quality being difficult to support the training of large models, high research and development costs, and poor algorithm autonomy and control. The application of large models should be guided by the needs of oil and gas business, taking the application of large models as an opportunity to improve data lifecycle management, enhance data governance capabilities, promote the construction of computing power, strengthen the construction of "artificial intelligence + energy" composite teams, and boost the autonomy and control of large model technology. [ABSTRACT FROM AUTHOR]

This study aimed to develop and evaluate a CT-based deep learning radiomics model for differentiating between Crohn's disease (CD) and intestinal tuberculosis (ITB). A total of 330 patients with pathologically confirmed as CD or ITB from the First Affiliated Hospital of Zhengzhou University were divided into the validation dataset one (CD: 167; ITB: 57) and validation dataset two (CD: 78; ITB: 28). Based on the validation dataset one, the synthetic minority oversampling technique (SMOTE) was adopted to create balanced dataset as training data for feature selection and model construction. The handcrafted and deep learning (DL) radiomics features were extracted from the arterial and venous phases images, respectively. The interobserver consistency analysis, Spearman's correlation, univariate analysis, and the least absolute shrinkage and selection operator (LASSO) regression were used to select features. Based on extracted multi-phase radiomics features, six logistic regression models were finally constructed. The diagnostic performances of different models were compared using ROC analysis and Delong test. The arterial-venous combined deep learning radiomics model for differentiating between CD and ITB showed a high prediction quality with AUCs of 0.885, 0.877, and 0.800 in SMOTE dataset, validation dataset one, and validation dataset two, respectively. Moreover, the deep learning radiomics model outperformed the handcrafted radiomics model in same phase images. In validation dataset one, the Delong test results indicated that there was a significant difference in the AUC of the arterial models (p = 0.037), while not in venous and arterial-venous combined models (p = 0.398 and p = 0.265) as comparing deep learning radiomics models and handcrafted radiomics models. In our study, the arterial-venous combined model based on deep learning radiomics analysis exhibited good performance in differentiating between CD and ITB. [ABSTRACT FROM AUTHOR]

Aim: The accessibility of social media data has allowed researchers to measure official–public interactions during COVID-19. However, previous work analyzing official posts or public comments has failed to explore the link between the two. Therefore, this study investigates the relationship between the communication strategies of public health agencies (PHAs) on TikTok and public emotional/sentiment tendencies in COVID-19 normalization. Subject and methods: This study uses the 2022 Shanghai city closure event as a public health communication case study in the context of COVID-19 normalization, using TikTok as a data source. We first analyze the communication strategies adopted by the PHA based on the Crisis and Emergency Risk Communication (CERC) model. Then, we classify the sentiment of public comments using the Large-Scale Knowledge Enhanced Pre-Training for Language Understanding and Generation (ERNIE) pre-training model. Finally, we explore the connection between PHA communication strategies and public sentiment tendencies. Results: First, the public's sentiment tendencies differ at different stages. Therefore, appropriate communication strategies should be developed stage-by-stage. Second, the public's emotional disposition to different communication strategies varies: government statements, vaccines, and prevention and control programs are more likely to produce a friendly comment environment, while policy and new cases per day are more likely to produce unfavorable comment content. However, this does not mean that policy and new cases per day should be avoided; the judicious use of these two strategies can help PHAs understand the current issues causing public dissatisfaction. Third, videos with celebrity appearances can significantly increase positive public sentiment and, thereby, public participation. Conclusion: We propose an improved CERC guideline for China based on the Shanghai lockdown case. [ABSTRACT FROM AUTHOR]

Data-driven prospectivity modeling based on deep learning, particularly supervised learning, has demonstrated outstanding performance for mineral exploration targeting in the past years, thanks to its powerful feature learning ability. However, this approach necessitates a substantial amount of large, high-quality labeled training data, and the scarcity of known mineral deposits poses significant challenges in constructing a high-performance mineral prospectivity prediction model. Self-supervised contrastive learning can alleviate this problem by exploiting large amounts of readily available unlabeled data. In this study, we utilized geochemical element data from the Malanyu district to train a self-supervised contrastive learning model. This model was then employed to predict gold mineral prospectivity, and its accuracy was compared with supervised learning method. The results show that the self-supervised contrastive learning model has higher performance in prospectivity prediction than the supervised learning model and its recognition accuracy reaches 100.00%, which is 7.41% higher than that of the supervised learning model ResNet50 and 14.81% higher than that of the supervised learning model MobileNetV2. At the same time, the prediction results of gold prospecting have a strong consistency with the known gold deposits in this district. This study demonstrates the feasibility of applying the self-supervised comparative learning model to the prediction of gold prospects, and it is of great significance to realize intelligent prediction of mineral resources. Article Highlights: A self-supervised contrastive learning model is proposed for predicting gold prospectivity. The model can learn valuable feature representations from a large amount of readily available unlabeled geochemical data. Compared to the supervised learning method, the proposed model not only reduces reliance on the number of known deposits but also achieves higher accuracy. [ABSTRACT FROM AUTHOR]

Intelligent forecasting of economic indexes has been an important demand for sustainable management of smart cities. Existing methods for this purpose were mostly established upon the basis of economic mechanism. Econometric models are the most general technical means in this area. However, in era of digital economy, increasing amount of big data has brought great change to traditional production. It is becoming more difficult for conventional technological forecasting methods to deal with multi-dimensional economic indexes. To deal with such challenge, this paper introduces the artificial intelligence algorithms to implement automatic information processing, and proposes a deep neural network-based intelligent forecasting method for multi-dimensional economic indexes in smart cities. Specifically, a deep neural network with three-layer structure is developed as the backbone methodology. For empirical analysis, the real-world data from "Chengdu–Chongqing Economic Circle" in China from 2012 to 2022 are selected as the main simulation scenario. Four major indexes are selected as the main research object: gross product (GDP), per capita GDP, GDP growth rate and the proportion of tertiary industry in GDP. The experimental results show that the proposal can well deal with such forecasting problem from a data-driven perspective, with a proper forecasting effect on historical data. [ABSTRACT FROM AUTHOR]

Project management requires the chaotic but important task of estimating software development effort. Several soft computing approaches have been proposed to increase estimation accuracy, and optimization techniques are utilized to concentrate on key aspects. However, a majority of works use data processing that has been found to be unreliable, time-consuming and typically leads to greater error rates. Therefore, the research proposes an efficient software development effort prediction model employing unique deep learning technology to alleviate the existing limitations. Data collection, pre-processing, feature selection and software development effort prediction are only a few of the varied phases of the proposed objective. After data collection, the data are pre-processed, including data cleaning, normalization, missing values and imputation. The expanded archer fish optimization method (Ext_AFO) is used to choose the best features from the pre-processed data. The Multilayer Perceptron Assisted Honey Bidirectional Gated Recurrent Feed Forward Network (Multi-HBiG) is built into this research work to provide an intelligent prediction model for software development effort estimation. The model parameters are adjusted using the Adaptive Honey Badger Optimisation Algorithm (A-Hba) to improve the overall estimation performance. The Albrecht dataset, China dataset, Desharnais dataset, Kemerer dataset, Maxwell dataset, Kitchenham dataset and Cocomos81 dataset are the datasets used in this study. The proposed approach surpassed other models when compared against Mean Relative Error (MRE), Mean Magnitude of Relative Error (MMRE), Mean Balanced Relative Error (MBRE) and Mean Inverted Balanced Relative Error (MIBRE) in the results section. The proposed model was assessed in this study using the MAE in each dataset, and it achieved 0.0753 for the China dataset, 0.0763 for the Cocomos81 dataset, 0.0737 for the Desharnais dataset, 0.0754 for the Kemerer dataset, 0.0759 for the Kitchenham dataset, 0.0734 for the Maxwell dataset and 0.0737 for the Albrecht dataset, respectively. [ABSTRACT FROM AUTHOR]

The joint action of human activities and environmental changes contributes to the frequent occurrence of landslide, causing major hazards. Using Interferometric Synthetic Aperture Radar (InSAR) technique enables the detailed detection of surface deformation, facilitating early landslide detection. The growing availability of SAR data and the development of artificial intelligence have spurred the integration of deep learning methods with InSAR for intelligent geological identification. However, existing studies using deep learning methods to detect landslides in InSAR deformation often rely on single InSAR data, which leads to the presence of other types of geological hazards in the identification results and limits the accuracy of landslide identification. Landslides are affected by many factors, especially topographic features. To enhance the accuracy of landslide identification, this study improves the existing geological hazard detection model and proposes a multi-source data fusion network termed MSFD-Net. MSFD-Net employs a pseudo-Siamese network without weight sharing, enabling the extraction of texture features from the wrapped deformation data and topographic features from topographic data, which are then fused in higher-level feature layers. We conducted comparative experiments on different networks and ablation experiments, and the results show that the proposed method achieved the best performance. We applied our method to the middle and upper reaches of the Yellow River in eastern Qinghai Province, China, and obtained deformation rates using Sentinel-1 SAR data from 2018 to 2020 in the region, ultimately identifying 254 landslides. Quantitative evaluations reveal that most detected landslides in the study area occurred at an elevation of 2500–3700 m with slope angles of 10–30°. The proposed landslide detection algorithm holds significant promise for quickly and accurately detecting wide-area landslides, facilitating timely preventive and control measures. [ABSTRACT FROM AUTHOR]

Water conservation has become a global problem as the population increases. In many densely populated cities in China, leaks from century-old pipe works have been widespread. However, entirely eradicating the issues involves replacing all water networks, which is costly and time-consuming. This paper proposed an AI-enabled water-saving control system with three control modes: time division control, flow regulation, and critical point control according to actual flow. Firstly, based on the current leaking situation of water supply networks in China and the capability level of China's water management, a water-saving technology integrating PID control and a series of deep learning algorithms was proposed. Secondly, a multi-jet control valve was designed to control pressure and reduce water distribution network cavitation. This technology has been successfully applied in industrial settings in China and has achieved gratifying water-saving results. [ABSTRACT FROM AUTHOR]

In recent years, the State Grid of China has placed significant emphasis on the monitoring of noise in substations, driven by growing environmental concerns. This paper presents a substation noise monitoring system designed based on an end-network-cloud architecture, aiming to acquire and analyze substation noise, and report anomalous noise levels that exceed national standards for substation operation and maintenance. To collect real-time noise data at substations, a self-developed noise acquisition device is developed, enabling precise analysis of acoustic characteristics. Moreover, to subtract the interfering environmental background noise (bird/insect chirping, human voice, etc.) and determine if noise exceedances are originating from substation equipment, an intelligent noise separation algorithm is proposed by leveraging the convolutional time-domain audio separation network (Conv-TasNet), dual-path recurrent neural network (DPRNN), and dual-path transformer network (DPTNet), respectively, and evaluated under various scenarios. Experimental results show that (1) deep-learning-based separation algorithms outperform the traditional spectral subtraction method, where the signal-to-distortion ratio improvement (SDRi) and the scale-invariant signal-to-noise ratio improvement (SI-SNRi) of Conv-TasNet, DPRNN, DPTNet and the traditional spectral subtraction are 12.6 and 11.8, 13.6 and 12.4, 14.2 and 12.9, and 4.6 and 4.1, respectively; (2) DPTNet and DPRNN exhibit superior performance in environment noise separation and substation equipment noise separation, respectively; and (3) 91% of post-separation data maintains sound pressure level deviations within 1 dB, showcasing the effectiveness of the proposed algorithm in separating interfering noises while preserving the accuracy of substation noise sound pressure levels. [ABSTRACT FROM AUTHOR]

The outbreaks of green tide have caused severe harm to the marine environment and human society. Synthetic Aperture Radar (SAR) plays an important role in green tide monitoring by virtue of its high resolution and cloud-free nature. The existing green tide extraction methods still face challenges in identifying multi-scale green tide patches due to noise interference, uneven greyscale and blurred boundaries in SAR images. Meanwhile, the practical application of deep learning methods with high precision is limited due to the complexity of the model and the large amount of computation. Therefore, we propose a multi-scale context-aware and batch-independent lightweight green tide extraction network called MBL-Net. A novel lightweight heterogeneous backbone is designed to extract multi-scale discriminative features and improve segmentation efficiency by using multi-scale selection kernel (MSK) modules and lightweight stages. Meanwhile, Triplet attention module is introduced to improve the internal consistency of the green tide region and suppress the effect of speckle noise. Then, the mixed pooling-based channel prior module (MCPM) is used to expand the receptive field of the network and extract the fine green tide structure by fusing multi-scale features. In addition, Filter Response Normalisation (FRN) is innovatively applied for feature normalization in the decoding stage, eliminating batch dependency. In order to verify the effectiveness of the proposed method, a dataset is built using the Sentinel-1 images of the Yellow Sea, China, from 2019 to 2021. The experimental results show that the proposed method achieves an overall accuracy of 98.59% with 0.970 G FLOPs and 3.525 M parameters, which ensures high precision and improves green tide detection efficiency. Compared with several representative networks, this method can capture more details of green tide with fewer parameters and faster calculation speed. [ABSTRACT FROM AUTHOR]

Classification of cryptic species is important for assessing biodiversity and conducting ecological studies. However, morphological classification methods face the loss of morphological information due to subjectivity in geometric morphometrics, while an incomplete database and horizontal gene transfer limit the molecular approach. A novel approach combining 3D modeling and artificial intelligence algorithms using morphological and molecular data was developed for species classification. Cryptic species from the Vignadula genus were used to test the feasibility of this new approach. Molecular identification results as data labels were used for training models, and for validating classification results of machine learning and deep learning. Our approach achieved accuracies of over 80% in distinguishing between V. atrata and V. mangle, which were identified by molecular data along China's coast. The result of the confusion matrix indicated the misidentified individuals were due to the morphological similarity in the intermediate zone. The feature importance analysis highlighted the significant contribution of average curvature—a 3D feature—to the task, indicating the feasibility of the 3D model in cryptic species classification. Utilizing 3D models and artificial intelligence, this study presents a novel approach for classifying cryptic species of molluscs. [ABSTRACT FROM AUTHOR]

Nonlinear deformation is a dynamically changing pattern of multiple surface deformations caused by groundwater overexploitation, underground coal mining, landslides, urban construction, etc., which are often accompanied by severe damage to surface structures or lead to major geological disasters; therefore, the high-precision monitoring and prediction of nonlinear surface deformation is significant. Traditional deep learning methods encounter challenges such as long-term dependencies or difficulty capturing complex spatiotemporal patterns when predicting nonlinear deformations. In this study, we developed a dual-attention-mechanism CNN-LSTM network model (DACLnet) to monitor and accurately predict nonlinear surface deformations precisely. Using advanced time series InSAR results as input, the DACLnet integrates the spatial feature extraction capability of a convolutional neural network (CNN), the advantages of the time series learning of a long short-term memory (LSTM) network, and the enhanced focusing effect of the dual-attention mechanism on crucial information, significantly improving the prediction accuracy of nonlinear surface deformations. The groundwater overexploitation area of the Turpan Basin, China, is selected to test the nonlinear deformation prediction effect of the proposed DACLnet. The results demonstrate that the DACLnet accurately captures developmental trends in historical surface deformations and effectively predicts surface deformations for the next two months in the study area. Compared to traditional LSTM and CNN-LSTM methods, the root mean square error (RMSE) of the DACLnet improved by 85.09% and 68.57%, respectively. These research results can provide crucial technical support for the early warning and prevention of geological disasters and can serve as an effective alternative tool for short-term ground subsidence prediction in areas lacking hydrogeological and other related data. [ABSTRACT FROM AUTHOR]

Although atmospheric CO2 concentrations collected by satellites play a crucial role in understanding global greenhouse gases, the sparse geographic distribution greatly affects their widespread application. In this paper, a hybrid CNN and spatiotemporal Kriging (CNN-STK) model is proposed to generate a monthly spatiotemporal continuous XCO2 dataset over China at 0.25° grid-scale from 2015 to 2020, utilizing OCO-2 XCO2 and geographic covariates. The validations against observation samples, CAMS XCO2 and TCCON measurements indicate the CNN-STK model is effective, robust, and reliable with high accuracy (validation set metrics: R2 = 0.936, RMSE = 1.3 ppm, MAE = 0.946 ppm; compared with TCCON: R2 = 0.954, RMSE = 0.898 ppm and MAE = 0.741 ppm). The accuracy of CNN-STK XCO2 exhibits spatial inhomogeneity, with higher accuracy in northern China during spring, autumn, and winter and lower accuracy in northeast China during summer. XCO2 in low-value-clustering areas is notably influenced by biological activities. Moreover, relatively high uncertainties are observed in the Qinghai-Tibet Plateau and Sichuan Basin. This study innovatively integrates deep learning with the geostatistical method, providing a stable and cost-effective approach for other countries and regions to obtain regional scales of atmospheric CO2 concentrations, thereby supporting policy formulation and actions to address climate change. [ABSTRACT FROM AUTHOR]

Thunderstorm gusts are a common form of severe convective weather in the warm season in North China, and it is of great importance to correctly forecast them. At present, the forecasting of thunderstorm gusts is mainly based on traditional subjective methods, which fails to achieve high-resolution and high-frequency gridded forecasts based on multiple observation sources. In this paper, we propose a deep learning method called Thunderstorm Gusts TransU-net (TG-TransUnet) to forecast thunderstorm gusts in North China based on multi-source gridded product data from the Institute of Urban Meteorology (IUM) with a lead time of 1 to 6 h. To determine the specific range of thunderstorm gusts, we combine three meteorological variables: radar reflectivity factor, lightning location, and 1-h maximum instantaneous wind speed from automatic weather stations (AWSs), and obtain a reasonable ground truth of thunderstorm gusts. Then, we transform the forecasting problem into an image-to-image problem in deep learning under the TG-TransUnet architecture, which is based on convolutional neural networks and a transformer. The analysis and forecast data of the enriched multi-source gridded comprehensive forecasting system for the period 2021–23 are then used as training, validation, and testing datasets. Finally, the performance of TG-TransUnet is compared with other methods. The results show that TG-TransUnet has the best prediction results at 1–6 h. The IUM is currently using this model to support the forecasting of thunderstorm gusts in North China. [ABSTRACT FROM AUTHOR]

The article examines the impact of AI tools in extracurricular online dance classes on student learning outcomes. The approach to interactive dance learning within Choreography (which is the mandatory discipline) using innovative M-learning platforms and apps such as Moodle and STEEZY has been introduced. An educational experiment was conducted among 40 students of the Institute of Music and Choreography at Ningxia Pedagogical University. The dance accomplishments in the control and experimental groups were assessed using the Choreographic Creativity Rating Scale in three areas: physical skills, presentation, and creativity. The mean levels of dancers' choreographic skills, as assessed by experts and audience at the end of the educational experiment, were determined. Students' projects were presented in such directions as: Hip-hop, Open Style, K-pop, House, Breaking, Popping, Whacking, Krump, Jazz Funk. The assessment of 4 levels of dance choreography (level 1—below expectations, level 2—meets some expectations, level 3—fully meets expectations, and level 4—exceeds expectations) in the areas of physical skills, presentation, and creativity of dancers' skills made it possible to compare learning outcomes in the control group and the experimental group. The expert assessment of students' achievements suggested that additional online extracurricular activities contribute to better dance skills, effective development of dancers' physical (+1.6), presentation (+1.16) and creative (+1.01) skills. This article is intended for dance instructors developing effective courses using relevant digital tools. [ABSTRACT FROM AUTHOR]

Tree detection and canopy area measurement are important and difficult tasks in forest inventory, which are important for understanding forest stand structure. This study utilized remotely piloted aircraft (RPA) aerial photography technology to collect remote sensing images of forests in Xiong County, China, creating a dataset comprising 1200 images of six tree species. Based on this dataset, the paper proposes an optimized model, Att-Mask R-CNN, for canopy detection and segmentation. Att-Mask R-CNN outperforms the original models (Mask R-CNN and MS R-CNN) by achieving 65.29% mean average precision for detection, 80.44% mean intersection over union for segmentation, and 90.67% overall recognition rate for the six tree species. In addition, a pixel statistics method based on segmentation masks is introduced for estimating the vertical projected area of individual tree crowns, and comparisons between the measured and predicted vertical projected area of the crowns of six tree species (100 trees of each class) show an overall goodness-of-fit R2 of 85% and a relative root-mean-square error rRMSE of 12.81%. By using remote sensing images from RPAs and optimizing existing deep learning models, the detection and segmentation of individual tree canopies can be achieved, resulting in a more accurate understanding of forest structure, which provides scientific support for forest management and resource monitoring. [ABSTRACT FROM AUTHOR]

Pollution sources release contaminants into water bodies via sewage outfalls (SOs). Using high-resolution images to interpret SOs is laborious and expensive because it needs specific knowledge and must be done by hand. Integrating unmanned aerial vehicles (UAVs) and deep learning technology could assist in constructing an automated effluent SOs detection tool by gaining specialized knowledge. Achieving this objective requires high-quality image datasets for model training and testing. However, there is no satisfactory dataset of SOs. This study presents a high-quality dataset named the images for sewage outfalls objective detection (iSOOD). The 10481 images in iSOOD were captured using UAVs and handheld cameras by individuals from the river basin in China. This study has carefully annotated these images to ensure accuracy and consistency. The iSOOD has undergone technical validation utilizing the YOLOv10 series objective detection model. Our study could provide high-quality SOs datasets for enhancing deep-learning models with UAVs to achieve efficient and intelligent river basin management. [ABSTRACT FROM AUTHOR]

Background: The metastatic vascular patterns of hepatocellular carcinoma (HCC) are mainly microvascular invasion (MVI) and vessels encapsulating tumor clusters (VETC). However, most existing VETC‐related radiological studies still focus on the prediction of VETC status. Purpose: This study aimed to build and compare VETC‐MVI related models (clinical, radiomics, and deep learning) associated with recurrence‐free survival of HCC patients. Study Type: Retrospective. Population: 398 HCC patients (349 male, 49 female; median age 51.7 years, and age range: 22–80 years) who underwent resection from five hospitals in China. The patients were randomly divided into training cohort (n = 358) and test cohort (n = 40). Field Strength/Sequence: 3‐T, pre‐contrast T1‐weighted imaging spoiled gradient recalled echo (T1WI SPGR), T2‐weighted imaging fast spin echo (T2WI FSE), and contrast enhanced arterial phase (AP), delay phase (DP). Assessment: Two radiologists performed the segmentation of HCC on T1WI, T2WI, AP, and DP images, from which radiomic features were extracted. The RFS related clinical characteristics (VETC, MVI, Barcelona stage, tumor maximum diameter, and alpha fetoprotein) and radiomic features were used to build the clinical model, clinical‐radiomic (CR) nomogram, deep learning model. The follow‐up process was done 1 month after resection, and every 3 months subsequently. The RFS was defined as the date of resection to the date of recurrence confirmed by radiology or the last follow‐up. Patients were followed up until December 31, 2022. Statistical Tests: Univariate COX regression, least absolute shrinkage and selection operator (LASSO), Kaplan–Meier curves, log‐rank test, C‐index, and area under the curve (AUC). P < 0.05 was considered statistically significant. Results: The C‐index of deep learning model achieved 0.830 in test cohort compared with CR nomogram (0.731), radiomic signature (0.707), and clinical model (0.702). The average RFS of the overall patients was 26.77 months (range 1–80 months). Data Conclusion: MR deep learning model based on VETC and MVI provides a potential tool for survival assessment. Evidence Level: 3 Technical Efficacy: Stage 3 [ABSTRACT FROM AUTHOR]

Objectives: We aim to develop a deep learning model based on a convolutional neural network (CNN) combined with a classification algorithm (CA) to assist dentists in quickly and accurately diagnosing the stage of periodontitis. Materials and methods: Periapical radiographs (PERs) and clinical data were collected. The CNNs including Alexnet, VGG16, and ResNet18 were trained on PER to establish the PER-CNN models for no periodontal bone loss (PBL) and PBL. The CAs including random forest (RF), support vector machine (SVM), naive Bayes (NB), logistic regression (LR), and k-nearest neighbor (KNN) were added to the PER-CNN model for control, stage I, stage II and stage III/IV periodontitis. Heat map was produced using a gradient-weighted class activation mapping method to visualize the regions of interest of the PER-Alexnet model. Clustering analysis was performed based on the ten PER-CNN scores and the clinical characteristics. Results: The accuracy of the PER-Alexnet and PER-VGG16 models with the higher performance was 0.872 and 0.853, respectively. The accuracy of the PER-Alexnet + RF model with the highest performance for control, stage I, stage II and stage III/IV was 0.968, 0.960, 0.835 and 0.842, respectively. Heat map showed that the regions of interest predicted by the model were periodontitis bone lesions. We found that age and smoking were significantly related to periodontitis based on the PER-Alexnet scores. Conclusion: The PER-Alexnet + RF model has reached high performance for whole-case periodontal diagnosis. The CNN models combined with CA can assist dentists in quickly and accurately diagnosing the stage of periodontitis. [ABSTRACT FROM AUTHOR]

Land surface hydrologic models adeptly capture crucial terrestrial processes with a high level of spatial detail. Typically, these models incorporate numerous uncertain, spatially varying parameters, the specification of which can profoundly impact the simulation capabilities. There is a longstanding tradition wherein parameter calibration has served as the conventional procedure to enhance model performance. However, calibrating distributed land surface hydrologic models presents a great challenge, often resulting in uneven spatial performance due to the compression of information inherent in model outputs and observations into a single‐value objective function. To address this problem, we propose a novel Generative Adversarial Network‐based Parameter Optimization (GAN‐PO) method. By leveraging a deep neural network to discern model spatial biases, we train a generative network to produce spatially coherent parameter fields, minimizing distinctions between simulations and observations. By leveraging neural network‐based surrogate models to make the physical model differentiable, we employ GAN‐PO to calibrate the Variable Infiltration Capacity (VIC) model against evapotranspiration (ET) over China's Huaihe basin. The results show that GAN‐PO can diminish errors in simulated ET derived from default parameters across nearly all grid cells within the study region, surpassing the conventional calibration approach based on the parameter regionalization technique. Ablation analysis indicates that relying solely on the traditional loss could lead to deteriorated model performance, underscoring the crucial role of the discriminator. Notably, due to the discriminator's explicit identification of model spatial biases, GAN‐PO excels in maintaining spatial consistency, outperforming the state‐of‐the‐art differentiable parameter learning (dPL) method in terms of model spatial performance. Key Points: A novel generative adversarial network‐based parameter estimation method is proposed to calibrate distributed land surface hydrologic modelsBy employing a discriminator to identify model spatial biases, this method contributes to effective and spatially coherent parameter estimationThis method can substantially reduce model simulated errors at grid scale and achieve consistent spatial performance [ABSTRACT FROM AUTHOR]

Background: Ameloblastoma, a common benign tumor found in the jaw bone, necessitates accurate localization and segmentation for effective diagnosis and treatment. However, the traditional manual segmentation method is plagued with inefficiencies and drawbacks. Hence, the implementation of an AI-based automatic segmentation approach is crucial to enhance clinical diagnosis and treatment procedures. Methods: We collected CT images from 79 patients diagnosed with ameloblastoma and employed a deep learning neural network model for training and testing purposes. Specifically, we utilized the Mask R-CNN neural network structure and implemented image preprocessing and enhancement techniques. During the testing phase, cross-validation methods were employed for evaluation, and the experimental results were verified using an external validation set. Finally, we obtained an additional dataset comprising 200 CT images of ameloblastoma from a different dental center to evaluate the model's generalization performance. Results: During extensive testing and evaluation, our model successfully demonstrated the capability to automatically segment ameloblastoma. The DICE index achieved an impressive value of 0.874. Moreover, when the IoU threshold ranged from 0.5 to 0.95, the model's AP was 0.741. For a specific IoU threshold of 0.5, the model achieved an AP of 0.914, and for another IoU threshold of 0.75, the AP was 0.826. Our validation using external data confirms the model's strong generalization performance. Conclusion: In this study, we successfully applied a neural network model based on deep learning that effectively performs automatic segmentation of ameloblastoma. The proposed method offers notable advantages in terms of efficiency, accuracy, and speed, rendering it a promising tool for clinical diagnosis and treatment. [ABSTRACT FROM AUTHOR]

Background: Given the high cost of endoscopy in gastric cancer (GC) screening, there is an urgent need to explore cost-effective methods for the large-scale prediction of precancerous lesions of gastric cancer (PLGC). We aim to construct a hierarchical artificial intelligence-based multimodal non-invasive method for pre-endoscopic risk screening, to provide tailored recommendations for endoscopy. Methods: From December 2022 to December 2023, a large-scale screening study was conducted in Fujian, China. Based on traditional Chinese medicine theory, we simultaneously collected tongue images and inquiry information from 1034 participants, considering the potential of these data for PLGC screening. Then, we introduced inquiry information for the first time, forming a multimodality artificial intelligence model to integrate tongue images and inquiry information for pre-endoscopic screening. Moreover, we validated this approach in another independent external validation cohort, comprising 143 participants from the China-Japan Friendship Hospital. Results: A multimodality artificial intelligence-assisted pre-endoscopic screening model based on tongue images and inquiry information (AITonguequiry) was constructed, adopting a hierarchical prediction strategy, achieving tailored endoscopic recommendations. Validation analysis revealed that the area under the curve (AUC) values of AITonguequiry were 0.74 for overall PLGC (95% confidence interval (CI) 0.71–0.76, p < 0.05) and 0.82 for high-risk PLGC (95% CI 0.82–0.83, p < 0.05), which were significantly and robustly better than those of the independent use of either tongue images or inquiry information alone. In addition, AITonguequiry has superior performance compared to existing PLGC screening methodologies, with the AUC value enhancing 45% in terms of PLGC screening (0.74 vs. 0.51, p < 0.05) and 52% in terms of high-risk PLGC screening (0.82 vs. 0.54, p < 0.05). In the independent external verification, the AUC values were 0.69 for PLGC and 0.76 for high-risk PLGC. Conclusion: Our AITonguequiry artificial intelligence model, for the first time, incorporates inquiry information and tongue images, leading to a higher precision and finer-grained pre-endoscopic screening of PLGC. This enhances patient screening efficiency and alleviates patient burden. [ABSTRACT FROM AUTHOR]

Tornadoes, as highly destructive weather events, require accurate detection for effective decision‐making. Traditional radar‐based tornado detection algorithms (TDA) face challenges with limited tornado feature extraction capabilities, leading to high false alarm rates and low detection probabilities. This study introduces the Multi‐Task Identification Network (MTI‐Net), leveraging Doppler radar data to enhance tornado recognition. MTI‐Net integrates tornado detection and estimation tasks to acquire comprehensive spatial and locational information. As part of MTI‐Net, we introduce a novel backbone network of Multi‐Head Convolutional Block (MHCB), which incorporates Spatial and Channel Attention Units (SAU and CAU). SAU optimizes local tornado feature extraction, while CAU reduces false alarms by enhancing dependencies among input variables. Experiments demonstrate the superiority of MTI‐Net over TDA, with a decrease in false alarm rates from 0.94 to 0.46 and an increase in hit rates from 0.23 to 0.81, highlighting the effectiveness of MTI‐Net in handling small‐scale tornado events. Plain Language Summary: Tornadoes, highly destructive small‐scale weather phenomena, demand accurate detection for informed decision‐making. Although meteorological radars are commonly utilized for tornado identification, current methods often suffer from false alarms or missed detections due to radar noise. In this study, we introduce the multi‐task learning‐based identification network (MTI‐Net), which not only enables tornado detection but also estimates tornado counts within radar data. We integrate Convolutional Neural Networks (CNNs) with Transformer techniques to enhance the model's ability to capture tornado information. CNNs detect local details using filters, while Transformers manage global connections through attention mechanisms. A series of experiments demonstrate significant improvements in tornado detection with MTI‐Net compared to traditional methods. Key Points: A tornado dataset with detailed radar features was created over China from 2017 to 2023Multi‐task learning was designed to simultaneously infer tornado detection and tornado number estimationIntegration of spatial and channel attention units can better extract tornado features from radar data [ABSTRACT FROM AUTHOR]

In this study, a tropical cyclone or typhoon rainfall forecast model based on Random Forest is developed to forecast the daily rainfall at 133 weather stations in China. The input factors to the model training process include rainfall observations during 1960–2018, typhoon information (position and intensity), station information (position and altitude), and properties of the western Pacific subtropical high. Model evaluation shows that besides the distance between a station and cyclone, the subtropical high properties are ranked very high in the model's feature importance, especially the subtropical ridgeline, and intensity. These aspects of the subtropical high influence the location and timing of typhoon landfall. The forecast model has a correlation coefficient of about 0.73, an Index of Agreement of nearly 0.8, and a mean bias of 1.28 mm based on the training dataset. Biases are consistently low, with both positive and negative signs, for target stations in the outer rainband (up to 1000 km, beyond which the model does not forecast) of typhoons. The range of biases is much larger for target stations in the inner-core (0–200 km) region. In this region, the model mostly overestimates (underestimates) the small (large) rain rates. Cases study of Typhoon Doksuri and Talim in 2023, as independent cases, shows the high performance of the model in forecasting the peak rain rates and timing of their occurrence of the two impactful typhoons. [ABSTRACT FROM AUTHOR]

In the globalization trend, China's cultural heritage is in danger of gradually disappearing. The protection and inheritance of these precious cultural resources has become a critical task. This paper focuses on the Miao batik culture in Guizhou Province, China, and explores the application of knowledge graphs, natural language processing, and deep learning techniques in the promotion and protection of batik culture. We propose a dual-channel mechanism that integrates semantic and visual information, aiming to connect batik pattern features with cultural connotations. First, we use natural language processing techniques to automatically extract batik-related entities and relationships from the literature, and construct and visualize a structured batik pattern knowledge graph. Based on this knowledge graph, users can textually search and understand the images, meanings, taboos, and other cultural information of specific patterns. Second, for the batik pattern classification, we propose an improved ResNet34 model. By embedding average pooling and convolutional operations into the residual blocks and introducing long-range residual connections, the classification performance is enhanced. By inputting pattern images into this model, their categories can be accurately identified, and then the underlying cultural connotations can be understood. Experimental results show that our model outperforms other mainstream models in evaluation metrics such as accuracy, precision, recall, and F1-score, achieving 94.46%, 94.47%, 93.62%, and 93.8%, respectively. This research provides new ideas for the digital protection of batik culture and demonstrates the great potential of artificial intelligence technology in cultural heritage protection. [ABSTRACT FROM AUTHOR]

Precisely assessing the vulnerability of landslides is essential for effective risk assessment. The findings from such assessments will undoubtedly be in high demand, providing a solid scientific foundation for a range of critical initiatives aimed at disaster prevention and control. In the research, authors set the ancient core district of Sichuan Moxi Ancient Town as the research object; they conduct and give the final result of the geological survey. Fault influences are commonly utilized as key markers for delineating strata in the field of stratigraphy, and the slope distance, slope angle, slope aspect, elevation, terrain undulation, plane curvature, profile curvature, mean curvature, relative elevation, land use type, surface roughness, water influence, distance of the catchment, cumulative water volume, and the Normalized Vegetation Index (NDVI) are used along roads to calculate annual rainfall. With the purpose of the establishment of the evaluation system, there are 17 factors selected in total. Through the landslide-susceptibility assessment by the coupled models of DNN-I-SVM and DNN-I-LR nine factors had been selected; it was found that the Area Under the Curve (AUC) value of the Receiver Operating Characteristic Curve (ROC) was high, and the accuracy of the model is relatively high. The coupler, DNN-I-LR, gives 0.875 of an evaluation accuracy of AUC, higher than DNN-I-SVM, which yielded 0.860. It is necessary to note that, in this region, compared to the DNN-I-SVM model, the DNN-I-LR coupling model has better fitting and prediction abilities. [ABSTRACT FROM AUTHOR]

Human pose estimation, a key area in computer vision, benefits various fields. The comparative study of research approaches in the United States and China, both leaders in this domain, is vital for understanding and influencing global trends in this technology. This review collected 191 influential papers from 2014 to 2022, sourced from Google Scholar. The Structural Topic Model (STM) was utilized to analyze research content, preferences, and trends in research topics. Specifically, 10 topics were summarized, and topic proportions, preferences, intensities, and word clouds were displayed via visualization methods. The findings revealed: 1) research on feature extraction and depth image constituted the largest proportion, approximately 12.2%, while data training research accounted for the lowest proportion, around 7.9%; 2) the United States and China exhibited distinct research preferences: the United States focused more on model and data research, while China emphasized deep learning and neural networks; 3) both countries exhibited similar research trends within the same topics, and research on deep learning technologies has experienced a slowdown in recent years. By comparative study, this review offers valuable insights and guidance for future investigations and applications in human pose estimation, such as improving the quality and diversity of data sets. [ABSTRACT FROM AUTHOR]

In the development of meteorological detection technology and services, weather radar undoubtedly plays a pivotal role, especially in the monitoring and early warning of severe convective weather events, where it serves an irreplaceable function. This research delves into the landscape of weather radar research from 1945 to 2024, employing scientometric methods to investigate 13,981 publications from the Web of Science (WoS) core collection database. This study aims to unravel, for the first time, the foundational structures shaping the knowledge domain of weather radar over an 80-year period, exploring general features, collaboration, co-citation, and keyword co-occurrence. Key findings reveal a significant surge in both publications and citations post-1990, peaking in 2022 with 1083 publications and 13832 citations, signaling sustained growth and interest in the field after a period of stagnation. The United States, China, and European countries emerge as key drivers of weather radar research, with robust international collaboration playing a pivotal role in the field's rapid evolution. Analysis uncovers 30 distinct co-citation clusters, showcasing the progression of weather radar knowledge structures. Notably, deep learning emerges as a dynamic cluster, garnering attention and yielding substantial outcomes in contemporary research efforts. Over eight decades, the focus of weather radar investigations has transitioned from hardware and software enhancements to Artificial Intelligence (AI) technology integration and multifunctional applications across diverse scenarios. This study identifies four key areas for future research: leveraging AI technology, advancing all-weather observation techniques, enhancing system refinement, and fostering networked collaborative observation technologies. This research endeavors to support academics by offering an in-depth comprehension of the progression of weather radar research. The findings can be a valuable resource for scholars in efficiently locating pertinent publications and journals. Furthermore, policymakers can rely on the insights gleaned from this study as a well-organized reference point. [ABSTRACT FROM AUTHOR]

The frequent occurrence of landslides poses a serious threat to people's lives and property. In order to evaluate disaster hazards based on remote sensing images via machine learning means, it is essential to establish an image database with manually labeled boundaries of landslides. However, the existing datasets do not cover diverse types of mountainous landslides. To address this issue, we propose a high-resolution (1 m) diverse mountainous landslide remote sensing dataset (DMLD), including 990 landslide instances across different terrain in southwestern China. To evaluate the performance of the DMLD, seven state-of-the-art deep learning models with different loss functions were implemented on it. The experiment results demonstrate not only that all of these deep learning methods with different characteristics can adapt well to the DMLD, but also that the DMLD has potential adaptability to other geographical regions. [ABSTRACT FROM AUTHOR]

The Yellow River Delta wetlands in China belong to the coastal wetland ecosystem, which is one of the youngest and most characteristic wetlands in the world. The Yellow River Delta wetlands are constantly changed by inland sediment and the influence of waves and storm surges, so the accurate classification of the coastal wetlands in the Yellow River Delta is of great significance for the rational utilization, development and protection of wetland resources. In this study, the Yellow River Delta sentinel-2 multispectral data were processed by super-resolution synthesis, and the feature bands were optimized. The optimal feature-band combination scheme was screened using the OIF algorithm. A deep learning model attention mechanism ResNet based on feature optimization with attention mechanism integration into the ResNet network is proposed. Compared with the classical machine learning model, the AM_ResNet model can effectively improve the classification accuracy of the wetlands in the Yellow River Delta. The overall accuracy was 94.61% with a Kappa of 0.93, and they were improved by about 6.99% and 0.1, respectively, compared with the best-performing Random Forest Classification in machine learning. The results show that the method can effectively improve the classification accuracy of the wetlands in the Yellow River Delta. [ABSTRACT FROM AUTHOR]

Before 2008, China lacked high-coverage regional surface observation data, making it difficult for the China Meteorological Administration Land Data Assimilation System (CLDAS) to directly backtrack high-resolution, high-quality land assimilation products. To address this issue, this paper proposes a deep learning model named UNET_DCA, based on the UNET architecture, which incorporates a Dual Cross-Attention module (DCA) for multiscale feature fusion by introducing Channel Cross-Attention (CCA) and Spatial Cross-Attention (SCA) mechanisms. This model focuses on the near-surface 10-m wind field and achieves spatial downscaling from 6.25 km to 1 km. We conducted training and validation using data from 2020–2021, tested with data from 2019, and performed ablation experiments to validate the effectiveness of each module. We compared the results with traditional bilinear interpolation methods and the SNCA-CLDASSD model. The experimental results show that the UNET-based model outperforms SNCA-CLDASSD, indicating that the UNET-based model captures richer information in wind field downscaling compared to SNCA-CLDASSD, which relies on sequentially stacked CNN convolution modules. UNET_CCA and UNET_SCA, incorporating cross-attention mechanisms, outperform UNET without attention mechanisms. Furthermore, UNET_DCA, incorporating both Channel Cross-Attention and Spatial Cross-Attention mechanisms, outperforms UNET_CCA and UNET_SCA, which only incorporate one attention mechanism. UNET_DCA performs best on the RMSE, MAE, and COR metrics (0.40 m/s, 0.28 m/s, 0.93), while UNET_DCA_ars, incorporating more auxiliary information, performs best on the PSNR and SSIM metrics (29.006, 0.880). Evaluation across different methods indicates that the optimal model performs best in valleys, followed by mountains, and worst in plains; it performs worse during the day and better at night; and as wind speed levels increase, accuracy decreases. Overall, among various downscaling methods, UNET_DCA and UNET_DCA_ars effectively reconstruct the spatial details of wind fields, providing a deeper exploration for the inversion of high-resolution historical meteorological grid data. [ABSTRACT FROM AUTHOR]

Reliable and timely prediction of the shield tunneling path is essential to avoid deviation and successfully complete a tunneling project. This study presents a reinforcement learning-based new optimal model for improving the forecasting accuracy of the shield tunneling path and alleviating shield drivers' over-reliance on their practical experience. This model integrates the Q-learning network with a metaheuristic gray wolf algorithm to explore and exploit the implicit information of the shield machine through Q-Table. The proposed method is applied to a field tunneling case with data collected from a real tunneling scenario in Tianjin City, China. The model is also evaluated using various numerical benchmark approaches and compared to a deep learning method. The results show that the proposed model produces an accurate prediction with a root-mean-square error of 0.539 and correlation coefficient of 0.925 for pitch values. A sensitivity analysis indicated that the thrust force and the buried depth have a significant influence on the prediction of shield tunneling trajectory. [ABSTRACT FROM AUTHOR]

In recent years, frequent chemical production safety incidents in China have been primarily attributed to dangerous behaviors by workers. Current monitoring methods predominantly rely on manual supervision, which is not only inefficient but also prone to errors in complex environments and with varying target scales, leading to missed or incorrect detections. To address this issue, we propose a deep learning-based object detection model, YOLO-GP. First, we utilize a grouped pointwise convolutional (GPConv) module of symmetric structure to facilitate information exchange and feature fusion in the channel dimension, thereby extracting more accurate feature representations. Building upon the YOLOv8n model, we integrate the symmetric structure convolutional GPConv module and design the dual-branch aggregation module (DAM) and Efficient Spatial Pyramid Pooling (ESPP) module to enhance the richness of gradient flow information and the capture of multi-scale features, respectively. Finally, we develop a channel feature enhancement network (CFE-Net) to strengthen inter-channel interactions, improving the model's performance in complex scenarios. Experimental results demonstrate that YOLO-GP achieves a 1.56% and 11.46% improvement in the mAP@.5:.95 metric on a custom dangerous behavior dataset and a public Construction Site Safety Image Dataset, respectively, compared to the baseline model. This highlights its superiority in dangerous behavior object detection tasks. Furthermore, the enhancement in model performance provides an effective solution for improving accuracy and robustness, promising significant practical applications. [ABSTRACT FROM AUTHOR]

Accurate rainfall measurement with a precise spatial and temporal resolution is essential for making informed decisions during disasters and conducting scientific studies, particularly in regions characterized by intricate terrain and limited coverage of automated weather stations. Retrieval of precipitation with satellite is currently the most effective means to obtain precipitation over large‐scale areas. The key to enhancing the accuracy of precipitation estimation and forecasting in regions with complex terrain lies in effectively integrating satellite data with topographic information. This paper introduces a deep learning approach called AttUnet_R_SFT that utilizes high temporal, spatial, and spectral resolution data obtained from the Fengyun 4A satellite, and incorporates the Deep Spatial Feature Transform (SFT) layer to incorporate geographical data for estimating half‐hourly precipitation in northeastern China. We assess it by compared to operational near‐real‐time satellite precipitation products demonstrated to be successful in estimating precipitation and baseline deep learning models. According to the experimental findings, the AttUnet_R_SFT model outperforms practical precipitation products and baseline deep learning models in both identifying and estimating precipitation. The main enhancement of the model performance is shown in the windward slope of the Greater Khingan Mountains as a result of the successful incorporation of geographical data. Hence, the suggested framework holds the capability to function as a superior and dependable satellite‐derived precipitation estimation solution in regions characterized by intricate terrain and infrequent rainfall. The findings of this study indicate that the utilization of deep learning algorithms for satellite precipitation estimation shows potential as a fruitful avenue for further research. Plain Language Summary: A deep learning model named AttUnet_R_SFT is proposed, which use high temporal, spatial and spectral resolution data from the Fengyun 4A satellite, and combines with the Deep Spatial Feature Transform (SFT) layer to input geographic information for half‐hourly precipitation estimation in the complex terrain region represented by northeast China. The model can provide a reference for improving the performance of precipitation estimation in areas with complex topography. Key Points: A deep‐learning model is proposed to effectively fuse satellite multispectral data of the Fengyun 4A satellite, with topographic informationHalf‐hourly precipitation is estimated with higher temporal resolution, which is closer to the operational needs of weather forecastingAs precipitation in the study area is a non‐high‐frequency event, data enhancement is attempted to use and obtain effective results [ABSTRACT FROM AUTHOR]

Satellite time series data, widely used for land cover classification, often contain missing values due to cloud contamination, which can negatively affect classification. Numerous strategies have been developed to reconstruct the missing values to produce regular time series for machine learning classifiers, among which the compositing followed by the linear interpolation is most widely used. However, the classification improvement of linear interpolation for land cover classification has not been examined. Recently developed deep learning models such as long short term memory (LSTM) and Transformer allow such examination as they can classify time series with missing values. In this study, we compared the time series composites with missing values (without linear interpolation) and the linearly interpolated time series composites (without missing values) for land cover classification. About 18 thousand Harmonized Landsat Sentinel-2 (HLS) images acquired over Amur River Basin of China (890,308 km2) in 2021 were composited to 14 16-day periods. Two time series composites were classified, i.e., (i) the 16-day composites without interpolation that have on average 15.35% 16-day periods with missing values and (ii) the linearly interpolated 16-day composites with no missing values. The classifications showed that (1) between classifications with and without linear interpolation there was < 0.2% overall accuracy differences for the bidirectional LSTM (Bi-LSTM) and < 0.5% for the Transformer both of which were smaller than model training randomness; and (2) the computation time can be saved using composites without linear interpolation. The findings suggested that it is unnecessary to use the time-consuming linear interpolation in Bi-LSTM and Transformer-based land cover classifications. The findings were confirmed by experiments for sensitivity to the number of cloud-free composites and to different classification legends using crop type classifications. It implied the linear interpolation algorithm cannot reconstruct reliable time series for land cover classifications and historical use of such method is more about mitigating the inability of traditional classifiers to handle missing values rather than improving classifications. Linear interpolation is not necessary for LSTM and Transformer with capability to handle missing values. The training datasets and developed codes in this study are made publicly available. [ABSTRACT FROM AUTHOR]

The complexity in stock index futures markets, influenced by the intricate interplay of human behavior, is characterized as nonlinearity and dynamism, contributing to significant uncertainty in long-term price forecasting. While machine learning models have demonstrated their efficacy in stock price forecasting, they rely solely on historical price data, which, given the inherent volatility and dynamic nature of financial markets, are insufficient to address the complexity and uncertainty in long-term forecasting due to the limited connection between historical and forecasting prices. This paper introduces a pioneering approach that integrates financial theory with advanced deep learning methods to enhance predictive accuracy and risk management in China's stock index futures market. The SF-Transformer model, combining spot-forward parity and the Transformer model, is proposed to improve forecasting accuracy across short and long-term horizons. Formulated upon the arbitrage-free futures pricing model, the spot-forward parity model offers variables such as stock index price, risk-free rate, and stock index dividend yield for forecasting. Our insight is that the mutual information generated by these variables has the potential to significantly reduce uncertainty in long-term forecasting. A case study on predicting major stock index futures prices in China demonstrates the superiority of the SF-Transformer model over models based on LSTM, MLP, and the stock index futures arbitrage-free pricing model, covering both short and long-term forecasting up to 28 days. Unlike existing machine learning models, the Transformer processes entire time series concurrently, leveraging its attention mechanism to discern intricate dependencies and capture long-range relationships, thereby offering a holistic understanding of time series data. An enhancement of mutual information is observed after introducing spot-forward parity in the forecasting. The variation of mutual information and ablation study results highlights the significant contributions of spot-forward parity, particularly to the long-term forecasting. Overall, these findings highlight the SF-Transformer model's efficacy in leveraging spot-forward parity for reducing uncertainty and advancing robust and comprehensive approaches in long-term stock index futures price forecasting. [ABSTRACT FROM AUTHOR]

Landslides, widespread and highly dangerous geological disasters, pose significant risks to humankind and the ecological environment. Consequently, predicting landslides is vital for disaster prevention and mitigation strategies. At present, the predominant methods for predicting landslide susceptibility are evolving from conventional machine learning techniques to deep learning approaches. At present, the predominant methods for predicting landslide susceptibility are evolving from conventional machine learning techniques to deep learning approaches. Prior studies have shown that in the context of landslide susceptibility, these models frequently underperform relative to tree-based machine learning algorithms. This shortcoming has restricted the application of deep learning in this domain. To overcome this challenge, this study presents the TabNet algorithm, which combines the interpretability and selective feature extraction of tree models with the representation learning and comprehensive training capabilities of neural network models. This paper explores the potential of employing the TabNet algorithm for landslide susceptibility analysis in China's WanZhou region and evaluates its performance against traditional machine learning techniques. The experimental data indicate that the TabNet algorithm achieves a recall score of 0.898 and an AUC of 0.915, demonstrating a generalization capability that is comparable to that of classical machine learning algorithms. [ABSTRACT FROM AUTHOR]

Deep learning methods have demonstrated remarkable success in recognizing geochemical anomalies for mineral exploration. Typically, these methods identify anomalies by reconstructing the geochemical background, which is marked by long-distance spatial variability, giving rise to long-range spatial dependencies within geochemical signals. However, current deep learning models for geochemical anomaly recognition face limitations in capturing intricate long-range spatial dependencies. Additionally, concerns emerge from the uncertainty associated with preprocessing in existing deep learning models, which involve generating interpolated images and topological graphs to represent the spatial structure of geochemical samples. In this paper, we present a novel end-to-end method for geochemical anomaly extraction based on the Transformer model. Our model utilizes self-attention mechanism to adequately capture both global and local interconnections among geochemical samples from a holistic perspective, enabling the reconstruction of geochemical background. Moreover, the self-attention mechanism allows the Transformer model to directly input free-form geochemical samples, eliminating the uncertainty associated with the employment of prior interpolation or graph generation typically required for geochemical samples. To align geochemical data with Transformer's architecture, we tailor a specialized data organization integrating learnable positional encoding and data masking. This enables the ingestion of entire geochemical data into the Transformer for anomaly recognition. Capitalizing on the flexibility afforded by the attention mechanism, we devise a contrastive loss for training, establishing a self-supervised learning scheme that enhances model generalizability for anomaly recognition. The proposed method is utilized to recognize geochemical anomalies related to Au mineralization in the northwest Jiaodong Peninsula, Eastern China. By comparison with anomalies identified by models of graph attention network and geographically weighted regression, it is demonstrated that the proposed method is more effective and geologically sound in identifying mineralization-associated anomalies. This superior performance in geochemical anomaly recognition is attributed to its ability to capture long-range dependencies within geochemical data. [ABSTRACT FROM AUTHOR]

The development of deep learning technique has granted firms with new opportunities to substantially improve their risk management strategies for sustainable growth. This paper introduces a novel deep learning‐based financial hedging (DL‐HE) strategy to leverage the salient ability of deep learning in extracting nonlinear features from complex high dimensional data, thus boosting the management of inventory risks arising from erratic commodity prices. Using real‐world data, we find that the average annualized economic benefit of the proposed strategy is at least 1.21 million CNY for a typical aluminum firm carrying an average level of inventory in China, as compared with those of the traditional hedging strategies. Further analysis reveals that such an economic benefit can largely be explained by the efficacy of the proposed DL‐HE strategy in terms of significantly improving return while still effectively controlling risk. Moreover, the superior of this strategy remains robust when extending to copper and zinc. [ABSTRACT FROM AUTHOR]

In this study, a Long Short-Term Memory (LSTM) network approach is employed to evaluate the prediction performance of PM2.5 in the Beijing–Tianjin–Hebei region (BTH). The proposed method is evaluated using the hourly air quality datasets from the China National Environmental Monitoring Center, European Center for Medium-range Weather Forecasts ERA5 (ECMWF-ERA5), and Multi-resolution Emission Inventory for China (MEIC) for the years 2016 and 2017. The predicted PM2.5 concentrations demonstrate a strong correlation with the observed values (R2 = 0.871–0.940) in the air quality dataset. Furthermore, the model exhibited the best performance in situations of heavy pollution (PM2.5 > 150 μg/m3) and during the winter season, with respective R2 values of 0.689 and 0.915. In addition, the influence of ECMWF-ERA5's hourly meteorological factors was assessed, and the results revealed regional heterogeneity on a large scale. Further evaluation was conducted by analyzing the chemical components of the MEIC inventory on the prediction performance. We concluded that the same temporal profile may not be suitable for addressing emission inventories in a large area with a deep learning method. [ABSTRACT FROM AUTHOR]

This mixed methods study investigated the change types and the underlying mechanisms of Chinese undergraduates' deep learning approach at a research-oriented university in eastern China. In Study 1, the deep learning approach of 273 freshmen was assessed using R-SPQ-2F at the beginning and end of a semester. The changes were categorized into three types: Type 1 (unchanged), Type 2 (increased), and Type 3 (decreased). Multiple logistic regression analysis was conducted to identify the influencing factors for each type. In Study 2, longitudinal qualitative interviews were conducted with 16 students to validate the findings from Study 1 and explore the formation mechanisms of their deep learning changes. The research revealed the changing trajectories of undergraduates' deep learning approach and highlighted the significant impact of family social and cultural capital, career goals, achievement goals, self-efficacy, and the external learning environment on these changes. The findings encourage universities to create conducive conditions that foster the enhancement of undergraduates' learning progress in higher education. [ABSTRACT FROM AUTHOR]

Storm surges are a common natural hazard in China's southern coastal area which usually cause a great loss of human life and financial damages. With the economic development and population concentration of coastal cities, storm surges may result in more impacts and damage in the future. Therefore, it is of vital importance to conduct risk assessment to identify high-risk areas and evaluate economic losses. However, quantitative study of storm surge risk assessment in undeveloped areas of China is difficult, since there is a lack of building character and damage assessment data. Aiming at the problem of data missing in undeveloped areas of China, this paper proposes a methodology for conducting storm surge risk assessment quantitatively based on deep learning and geographic information system (GIS) techniques. Five defined storm surge inundation scenarios with different typhoon return periods are simulated by the coupled FVCOM–SWAN (Finite Volume Coastal Ocean Model–Simulating WAves Nearshore) model, the reliability of which is validated using official measurements. Building footprints of the study area are extracted through the TransUNet deep learning model and remote sensing images, while building heights are obtained through unoccupied aerial vehicle (UAV) measurements. Subsequently, economic losses are quantitatively calculated by combining the adjusted depth–damage functions and overlaying an analysis of the buildings exposed to storm surge inundation. Zoning maps of the study area are provided to illustrate the risk levels according to economic losses. The quantitative risk assessment and zoning maps can help the government to provide storm surge disaster prevention measures and to optimize land use planning and thus to reduce potential economic losses in the coastal area. [ABSTRACT FROM AUTHOR]

The rise of online‐to‐offline (O2O) e‐commerce business has brought tremendous opportunities to the logistics industry. In the online‐to‐offline logistics business, it is essential to detect anomaly merchants with fraudulent shipping behaviours, such as sending other merchants' packages for profit with their low discounts. This can help reduce the financial losses of platforms and ensure a healthy environment. Existing anomaly detection studies have mainly focused on online fraud behaviour detection, such as fraudulent purchase and comment behaviours in e‐commerce. However, these methods are not suitable for anomaly merchant detection in logistics due to the more complex online and offline operation of package‐sending behaviours and the interpretable requirements of offline deployment in logistics. MultiDet, a semi‐supervised multi‐view fusion‐based Anomaly Detection framework in online‐to‐offline logistics is proposed, which consists of a basic version SemiDet and an attention‐enhanced multi‐view fusion model. In SemiDet, pair‐wise data augmentation is first conducted to promote model robustness and address the challenge of limited labelled anomaly instances. Then, SemiDet calculates the anomaly scoring of each merchant with an auto‐encoder framework. Considering the multi‐relationships among logistics merchants, a multi‐view attention fusion‐based anomaly detection network is further designed to capture merchants' mutual influences and improve the anomaly merchant detection performance. A post‐hoc perturbation‐based interpretation model is designed to output the importance of different views and ensure the trustworthiness of end‐to‐end anomaly detection. The framework based on an eight‐month real‐world dataset collected from one of the largest logistics platforms in China is evaluated, involving 6128 merchants and 16 million historical order consignor records in Beijing. Experimental results show that the proposed model outperforms other baselines in both AUC‐ROC and AUC‐PR metrics. [ABSTRACT FROM AUTHOR]