Search

Deep learning models have demonstrated potential in Condition-Based Monitoring (CBM) for rotating machinery, such as induction motors (IMs). However, their performance is significantly influenced by the size of the training dataset and the way signals are presented to the model. When trained on segmented signals over a fixed period, the model's accuracy can decline when tested on signals that differ from the training interval or are randomly sampled. Conversely, models utilizing data augmentation techniques exhibit better generalization to unseen conditions. This paper highlights the bias introduced by traditional training methods towards specific periodic waveform sampling and proposes a new method to augment phase current signals during training using a shifting window technique. This approach is considered as a practical approach for motor current augmentation and is shown to enhance classification accuracy and improved generalization when compared to existing techniques. [ABSTRACT FROM AUTHOR]

Magnetic field is one of the basic data for constructing a transformer digital twin. The finite element transient simulation takes a long time and cannot meet the real-time requirements of a digital twin. According to the nonlinear characteristics of the core and the timing characteristics of the magnetic field, this paper proposes a fast calculation method of the spatial magnetic field of the transformer, considering the nonlinear characteristics of the core. Firstly, based on the geometric and electrical parameters of the single-phase double-winding test transformer, the corresponding finite element simulation model is built. Secondly, the key parameters of the finite element model are parametrically scanned to obtain the nonlinear working condition data set of the test transformer. Finally, a deep learning network integrating a convolutional neural network (CNN) and a long short-term memory network (LSTM) is built to train the mapping relationship between winding voltage, current, and the spatial magnetic field so as to realize the rapid calculation of the transformer magnetic field. The results show that the calculation time of the deep learning model is greatly shortened compared with the finite element model, and the model calculation results are consistent with the experimental measurement results. [ABSTRACT FROM AUTHOR]

This paper introduces a novel convolutional neural network (CNN) architecture tailored for state of charge (SoC) estimation in battery management systems (BMS), accompanied by an advanced optimization technique to enhance training efficiency. The proposed CNN architecture comprises multiple one-dimensional convolutional (Conv1D) layers followed by batch normalization and one-dimensional max-pooling (MaxPooling1D) layers, culminating in dense layers for regression-based SoC prediction. To improve training effectiveness, we introduce an advanced dynamic k-decay learning rate scheduling method. This technique dynamically adjusts the learning rate during training, responding to changes in validation loss to fine-tune the training process. Experimental validation was conducted on various drive cycles, including the dynamic stress test (DST), Federal Urban Driving Schedule (FUDS), Urban Dynamometer Driving Schedule (UDDS), United States 2006 Supplemental Federal Test Procedure (US06), and Worldwide Harmonized Light Vehicles Test Cycle (WLTC), spanning four temperature conditions (−5 °C, 5 °C, 25 °C, 45 °C). Notably, the test error of DST and US06 drive cycles, the CNN with optimization achieved a mean absolute error (MAE) of 0.0091 and 0.0080, respectively at 25 °C, and a root mean square error (RMSE) of 0.013 and 0.0095, respectively. In contrast, the baseline CNN without optimization yielded higher MAE and RMSE values of 0.011 and 0.014, respectively, on the same drive cycles. Additionally, training time with the optimization technique was significantly reduced, with a recorded time of 324.14 s compared to 648.59 s for the CNN without optimization at room temperature. These results demonstrate the effectiveness of the proposed CNN architecture combined with advanced dynamic learning rate scheduling in accurately predicting SoC across various battery types and drive cycles. The optimization technique not only improves prediction accuracy but also substantially reduces training time, highlighting its potential for enhancing battery management systems in electric vehicle applications. [ABSTRACT FROM AUTHOR]

Photovoltaic (PV) microgrids comprise a multitude of small PV power stations distributed across a specific geographical area in a decentralized manner. Computational services for forecasting the output power of power stations are crucial for optimizing resource deployment. This paper proposes a deep-learning-based architecture for short-term prediction of PV power. Firstly, in order to make full use of the spatial information between different power stations, a spatio–temporal feature fusion method is proposed. This method is capable of exploiting both the power information of neighboring power stations with strong correlations and meteorological information with the PV feature data of the target power station. By using a multiscale convolutional neural network–long short-term memory (CNN-LSTM) network model, it is capable of generating a PV feature dataset containing spatio–temporal attributes that expand the data source and enhance the feature constraints. It is capable of predicting the output power sequences of power stations in PV microgrids with high model generalization and responsiveness. To validate the effectiveness of the proposed framework, an extensive numerical analysis is also conducted based on a real-world PV dataset. [ABSTRACT FROM AUTHOR]

Seals, sea lions, and other aquatic animals rely on their whiskers to identify and track underwater targets, offering valuable inspiration for the development of low-power, portable, and environmentally friendly sensors. Here, we design a single seal-whisker-like cylinder and conduct experiments to measure the forces acting on it with nine different upstream targets. Using sample sets constructed from these force signals, a convolutional neural network (CNN) is trained and tested. The results demonstrate that combining the seal-whisker-style sensor with a CNN enables the identification of objects in the water in most cases, although there may be some confusion for certain targets. Increasing the length of the signal samples can enhance the results but may not eliminate these confusions. Our study reveals that high frequencies (greater than 5 Hz) are irrelevant in our model. Lift signals present more distinct and distinguishable features than drag signals, serving as the primary basis for the model to differentiate between various targets. Fourier analysis indicates that the model's efficacy in recognizing different targets relies heavily on the discrepancies in the spectral features of the lift signals. [ABSTRACT FROM AUTHOR]

Falls among the elderly are a common and serious health risk that can lead to physical injuries and other complications. To promptly detect and respond to fall events, radar-based fall detection systems have gained widespread attention. In this paper, a deep learning model is proposed based on the frequency spectrum of radar signals, called the convolutional bidirectional long short-term memory (CB-LSTM) model. The introduction of the CB-LSTM model enables the fall detection system to capture both temporal sequential and spatial features simultaneously, thereby enhancing the accuracy and reliability of the detection. Extensive comparison experiments demonstrate that our model achieves an accuracy of 98.83% in detecting falls, surpassing other relevant methods currently available. In summary, this study provides effective technical support using the frequency spectrum and deep learning methods to monitor falls among the elderly through the design and experimental validation of a radar-based fall detection system, which has great potential for improving quality of life for the elderly and providing timely rescue measures. [ABSTRACT FROM AUTHOR]

Tool condition monitoring technology is an indispensable part of intelligent manufacturing. Most current research focuses on complex signal processing techniques or advanced deep learning algorithms to improve prediction performance without fully leveraging the end-to-end advantages of deep learning. The challenge lies in transforming multi-sensor raw data into input data suitable for direct model feeding, all while minimizing data scale and preserving sufficient temporal interpretation of tool wear. However, there is no clear reference standard for this so far. In light of this, this paper innovatively explores the processing methods that transform raw data into input data for deep learning models, a process known as an input paradigm. This paper introduces three new input paradigms: the downsampling paradigm, the periodic paradigm, and the subsequence paradigm. Then an improved hybrid model that combines a convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM) was employed to validate the model's performance. The subsequence paradigm demonstrated considerable superiority in prediction results based on the PHM2010 dataset, as the newly generated time series maintained the integrity of the raw data. Further investigation revealed that, with 120 subsequences and the temporal indicator being the maximum value, the model's mean absolute error (MAE) and root mean square error (RMSE) were the lowest after threefold cross-validation, outperforming several classical and contemporary methods. The methods explored in this paper provide references for designing input data for deep learning models, helping to enhance the end-to-end potential of deep learning models, and promoting the industrial deployment and practical application of tool condition monitoring systems. [ABSTRACT FROM AUTHOR]

With the rapid advancement of the Internet of Things, network security has garnered increasing attention from researchers. Applying deep learning (DL) has significantly enhanced the performance of Network Intrusion Detection Systems (NIDSs). However, due to its complexity and "black box" problem, deploying DL-based NIDS models in practical scenarios poses several challenges, including model interpretability and being lightweight. Feature selection (FS) in DL models plays a crucial role in minimizing model parameters and decreasing computational overheads while enhancing NIDS performance. Hence, selecting effective features remains a pivotal concern for NIDSs. In light of this, this paper proposes an interpretable feature selection method for encrypted traffic intrusion detection based on SHAP and causality principles. This approach utilizes the results of model interpretation for feature selection to reduce feature count while ensuring model reliability. We evaluate and validate our proposed method on two public network traffic datasets, CICIDS2017 and NSL-KDD, employing both a CNN and a random forest (RF). Experimental results demonstrate superior performance achieved by our proposed method. [ABSTRACT FROM AUTHOR]

With the widespread use of unmanned aerial vehicles (UAVs), the detection and identification of UAVs is a vital security issue for the safety of airspace and ground facilities in the no-fly zone. Telemetry radios are important wireless communication devices for UAVs, especially in UAVs beyond the visual line of sight (BVLOS) operating mode. This work focuses on the UAV identification approach using transient signals from UAV telemetry radios instead of the signals from UAV controllers that the former research work depended on. In our novel UAV Radio Frequency (RF) identification system framework based on telemetry radio signals, the E C − α algorithm is optimized to detect the starting point of the UAV transient signal and the detection accuracy at different signal-to-noise ratios (SNR) is evaluated. In the training stage, the Convolutional Neural Network (CNN) model is trained to extract features from raw I/Q data of the transient signals with different waveforms. Its architecture and hyperparameters are analyzed and optimized. In the identification stage, the extracted transient signals are clustered through the Self-Organizing Map (SOM) algorithm and the Clustering Signals Joint Identification (CSJI) algorithm is proposed to improve the accuracy of RF fingerprint identification. To evaluate the performance of our proposed approach, we design a testbed, including two UAVs as the flight platform, a Universal Software Radio Peripheral (USRP) as the receiver, and 20 telemetry radios with the same model as targets for identification. Indoor test results show that the optimized identification approach achieves an average accuracy of 92.3% at 30 dB. In comparison, the identification accuracy of SVM and KNN is 69.7% and 74.5%, respectively, at the same SNR condition. Extensive experiments are conducted outdoors to demonstrate the feasibility of this approach. [ABSTRACT FROM AUTHOR]

Background: Chest X-rays (CXRs) are pivotal in clinical diagnostics, particularly in assessing cardiomegaly through the cardiothoracic ratio (CTR). This systematic review and meta-analysis evaluate the efficacy of artificial intelligence (AI) in automating CTR determination to enhance patient care and streamline diagnostic processes. They are concentrated on comparing the performance of AI models in determining the CTR against human assessments, identifying the most effective models for potential clinical implementation. This study was registered with PROSPERO (no. CRD42023437459). No funding was received. Methods: A comprehensive search of medical databases was conducted in June 2023. The search strategy adhered to the PICO framework. Inclusion criteria encompassed original articles from the last decade focusing on AI-assisted CTR assessment from standing-position CXRs. Exclusion criteria included systematic reviews, meta-analyses, conference abstracts, paediatric studies, non-original articles, and studies using imaging techniques other than X-rays. After initial screening, 117 articles were reviewed, with 14 studies meeting the final inclusion criteria. Data extraction was performed by three independent investigators, and quality assessment followed PRISMA 2020 guidelines, using tools such as the JBI Checklist, AMSTAR 2, and CASP Diagnostic Study Checklist. Risk of bias was assessed according to the Cochrane Handbook guidelines. Results: Fourteen studies, comprising a total of 70,472 CXR images, met the inclusion criteria. Various AI models were evaluated, with differences in dataset characteristics and AI technology used. Common preprocessing techniques included resizing and normalization. The pooled AUC for cardiomegaly detection was 0.959 (95% CI 0.944–0.975). The pooled standardized mean difference for CTR measurement was 0.0353 (95% CI 0.147–0.0760). Significant heterogeneity was found between studies (I2 89.97%, p < 0.0001), with no publication bias detected. Conclusions: Standardizing methodologies is crucial to avoid interpretational errors and advance AI in medical imaging diagnostics. Uniform reporting standards are essential for the further development of AI in CTR measurement and broader medical imaging applications. [ABSTRACT FROM AUTHOR]

This study investigated the usefulness of deep learning-based automatic detection of temporomandibular joint (TMJ) effusion using magnetic resonance imaging (MRI) in patients with temporomandibular disorder and whether the diagnostic accuracy of the model improved when patients' clinical information was provided in addition to MRI images. The sagittal MR images of 2948 TMJs were collected from 1017 women and 457 men (mean age 37.19 ± 18.64 years). The TMJ effusion diagnostic performances of three convolutional neural networks (scratch, fine-tuning, and freeze schemes) were compared with those of human experts based on areas under the curve (AUCs) and diagnosis accuracies. The fine-tuning model with proton density (PD) images showed acceptable prediction performance (AUC = 0.7895), and the from-scratch (0.6193) and freeze (0.6149) models showed lower performances (p < 0.05). The fine-tuning model had excellent specificity compared to the human experts (87.25% vs. 58.17%). However, the human experts were superior in sensitivity (80.00% vs. 57.43%) (all p < 0.001). In gradient-weighted class activation mapping (Grad-CAM) visualizations, the fine-tuning scheme focused more on effusion than on other structures of the TMJ, and the sparsity was higher than that of the from-scratch scheme (82.40% vs. 49.83%, p < 0.05). The Grad-CAM visualizations agreed with the model learned through important features in the TMJ area, particularly around the articular disc. Two fine-tuning models on PD and T2-weighted images showed that the diagnostic performance did not improve compared with using PD alone (p < 0.05). Diverse AUCs were observed across each group when the patients were divided according to age (0.7083–0.8375) and sex (male:0.7576, female:0.7083). The prediction accuracy of the ensemble model was higher than that of the human experts when all the data were used (74.21% vs. 67.71%, p < 0.05). A deep neural network (DNN) was developed to process multimodal data, including MRI and patient clinical data. Analysis of four age groups with the DNN model showed that the 41–60 age group had the best performance (AUC = 0.8258). The fine-tuning model and DNN were optimal for judging TMJ effusion and may be used to prevent true negative cases and aid in human diagnostic performance. Assistive automated diagnostic methods have the potential to increase clinicians' diagnostic accuracy. [ABSTRACT FROM AUTHOR]

In recent years, deep learning techniques for pansharpening multiresolution images have gained increasing interest. Due to the lack of ground truth data, most deep learning solutions rely on synthetic reduced-resolution data for supervised training. This approach has limitations due to the statistical mismatch between real full-resolution and synthetic reduced-resolution data, which affects the models' generalization capacity. Consequently, there has been a shift towards unsupervised learning frameworks for pansharpening deep learning-based techniques. Unsupervised schemes require defining sophisticated loss functions with at least two components: one for spectral quality, ensuring consistency between the pansharpened image and the input multispectral component, and another for spatial quality, ensuring consistency between the output and the panchromatic input. Despite promising results, there has been limited investigation into the interaction and balance of these loss terms to ensure stability and accuracy. This work explores how unsupervised spatial and spectral consistency losses can be reliably combined preserving the outocome quality. By examining these interactions, we propose a general rule for balancing the two loss components to enhance the stability and performance of unsupervised pansharpening models. Experiments on three state-of-the-art algorithms using WorldView-3 images demonstrate that methods trained with the proposed framework achieve good performance in terms of visual quality and numerical indexes. [ABSTRACT FROM AUTHOR]

The precise estimation of above-ground biomass (AGB) is imperative for the advancement of breeding programs. Optical variables, such as vegetation indices (VI), have been extensively employed in monitoring AGB. However, the limited robustness of inversion models remains a significant impediment to the widespread application of UAV-based multispectral remote sensing in AGB inversion. In this study, a novel stem–leaf separation strategy for AGB estimation is delineated. Convolutional neural network (CNN) and transfer learning (TL) methodologies are integrated to estimate leaf biomass (LGB) across multiple growth stages, followed by the development of an allometric growth model for estimating stem biomass (SGB). To enhance the precision of LGB inversion, the large-scale remote sensing data and image simulation framework over heterogeneous scenes (LESS) model, which is a three-dimensional (3D) radiative transfer model (RTM), was utilized to simulate a more extensive canopy spectral dataset, characterized by a broad distribution of canopy spectra. The CNN model was pre-trained in order to gain prior knowledge, and this knowledge was transferred to a re-trained model with a subset of field-observed samples. Finally, the allometric growth model was utilized to estimate SGB across various growth stages. To further validate the generalizability, transferability, and predictive capability of the proposed method, field samples from 2022 and 2023 were employed as target tasks. The results demonstrated that the 3D RTM + CNN + TL method outperformed best in LGB estimation, achieving an R² of 0.73 and an RMSE of 72.5 g/m² for the 2022 dataset, and an R² of 0.84 and an RMSE of 56.4 g/m² for the 2023 dataset. In contrast, the PROSAIL method yielded an R² of 0.45 and an RMSE of 134.55 g/m² for the 2022 dataset, and an R² of 0.74 and an RMSE of 61.84 g/m² for the 2023 dataset. The accuracy of LGB inversion was poor when using only field-measured samples to train a CNN model without simulated data, with R² values of 0.30 and 0.74. Overall, learning prior knowledge from the simulated dataset and transferring it to a new model significantly enhanced LGB estimation accuracy and model generalization. Additionally, the allometric growth model's estimation of SGB resulted in an accuracy of 0.87 and 120.87 g/m² for the 2022 dataset, and 0.74 and 86.87 g/m² for the 2023 dataset, exhibiting satisfactory results. Separate estimation of both LGB and SGB based on stem and leaf separation strategies yielded promising results. This method can be extended to the monitor and inversion of other critical variables. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) and graph convolutional networks (GCNs) have made considerable advances in hyperspectral image (HSI) classification. However, most CNN-based methods learn features at a single-scale in HSI data, which may be insufficient for multi-scale feature extraction in complex data scenes. To learn the relations among samples in non-grid data, GCNs are employed and combined with CNNs to process HSIs. Nevertheless, most methods based on CNN-GCN may overlook the integration of pixel-wise spectral signatures. In this paper, we propose a pyramid cascaded convolutional neural network with graph convolution (PCCGC) for hyperspectral image classification. It mainly comprises CNN-based and GCN-based subnetworks. Specifically, in the CNN-based subnetwork, a pyramid residual cascaded module and a pyramid convolution cascaded module are employed to extract multiscale spectral and spatial features separately, which can enhance the robustness of the proposed model. Furthermore, an adaptive feature-weighted fusion strategy is utilized to adaptively fuse multiscale spectral and spatial features. In the GCN-based subnetwork, a band selection network (BSNet) is used to learn the spectral signatures in the HSI using nonlinear inter-band dependencies. Then, the spectral-enhanced GCN module is utilized to extract and enhance the important features in the spectral matrix. Subsequently, a mutual-cooperative attention mechanism is constructed to align the spectral signatures between BSNet-based matrix with the spectral-enhanced GCN-based matrix for spectral signature integration. Abundant experiments performed on four widely used real HSI datasets show that our model achieves higher classification accuracy than the fourteen other comparative methods, which shows the superior classification performance of PCCGC over the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The vision transformer (ViT) has demonstrated performance comparable to that of convolutional neural networks (CNN) in the hyperspectral image classification domain. This is achieved by transforming images into sequence data and mining global spectral-spatial information to establish remote dependencies. Nevertheless, both the ViT and CNNs have their own limitations. For instance, a CNN is constrained by the extent of its receptive field, which prevents it from fully exploiting global spatial-spectral features. Conversely, the ViT is prone to excessive distraction during the feature extraction process. To be able to overcome the problem of insufficient feature information extraction caused using by a single paradigm, this paper proposes an MLP-mixer and a graph convolutional enhanced transformer (MGCET), whose network consists of a spatial-spectral extraction block (SSEB), an MLP-mixer, and a graph convolutional enhanced transformer (GCET). First, spatial-spectral features are extracted using SSEB, and then local spatial-spectral features are fused with global spatial-spectral features by the MLP-mixer. Finally, graph convolution is embedded in multi-head self-attention (MHSA) to mine spatial relationships and similarity between pixels, which further improves the modeling capability of the model. Correlation experiments were conducted on four different HSI datasets. The MGEET algorithm achieved overall accuracies (OAs) of 95.45%, 97.57%, 98.05%, and 98.52% on these datasets. [ABSTRACT FROM AUTHOR]

Lung cancer patients’ chances of survival can increase with early detection. For the diagnosis of lung cancer, precise lung nodule identification in computed tomography (CT) images is crucial. The variety of the lung nodules and the complexity of the surroundings, however, have made robust nodule identification a challenging problem. To address this challenge, the Extensible Hunt Optimization-based deep Convolutional Neural Network classifier (EHO-deep CNN), is developed in this research for detecting the lung nodules present in the CT images. The proposed model in which the preprocessing is performed through the median filter to eliminate the artifacts in the image. Specifically, the images are segmented by adopting the centralized superpixels segmentation-based iterative clustering (CSSBIC) technique that enhances the accuracy of segmentation. The proposed approach relies on the GLCM and the VGG16-based feature extraction to explicate the discrimination of the lung nodule features. In addition, the retrieved features are combined and fed forwarded to the Deep CNN classifier that is optimally tuned via the EHO algorithm. The EHO algorithm is developed by fusing the fitness function of Border collie dogs with the hunting characteristics of Dingo to enhance convergence rate and performance. The proposed EHO-deep CNN is reported in terms of accuracy, sensitivity, and specificity values as 98.13%, 98.98%, and 97.28% when measuring TP 80 for the LIDC-IDRI dataset and 96.09%, 96.70%, and 95.48% for measuring k-fold 10. Further, implementing the suggested model obtains the values of 95.34%, 92.88%, and 97.80% for the LUNA 16 dataset when measuring TP 80 and 98.21%, 98.06%, 98.35% when measuring k-fold 10, which outperforms other existing techniques in effectiveness. [ABSTRACT FROM AUTHOR]

Simple Summary: Accurate identification of individual Amur tigers is vital for their conservation, as it helps us understand their population and distribution. Existing identification methods often fall short in accuracy, and our study focuses on creating a more accurate method for identifying individual Amur tigers using advanced deep learning techniques. We improved an existing neural network model called InceptionResNetV2 by adding features like dropout layers and dual-attention mechanisms to better capture the unique stripe patterns of each tiger and reduce errors during training. We tested our model on a large dataset of tiger images and found it to be highly effective, achieving an average recognition accuracy of over 95% for different body parts, with left stripes reaching the highest 99.37%. This method significantly outperforms previous models and provides a reliable tool for wildlife researchers and conservationists to monitor and protect Amur tigers. By improving the ability to track individual tigers, our research offers practical benefits for preserving this endangered species and enhancing wildlife management practices globally. Accurate and intelligent identification of rare and endangered individuals of flagship wildlife species, such as Amur tiger (Panthera tigris altaica), is crucial for understanding population structure and distribution, thereby facilitating targeted conservation measures. However, many mathematical modeling methods, including deep learning models, often yield unsatisfactory results. This paper proposes an individual recognition method for Amur tigers based on an improved InceptionResNetV2 model. Initially, the YOLOv5 model is employed to automatically detect and segment facial, left stripe, and right stripe areas from images of 107 individual Amur tigers, achieving a high average classification accuracy of 97.3%. By introducing a dropout layer and a dual-attention mechanism, we enhance the InceptionResNetV2 model to better capture the stripe features of individual tigers at various granularities and reduce overfitting during training. Experimental results demonstrate that our model outperforms other classic models, offering optimal recognition accuracy and ideal loss changes. The average recognition accuracy for different body part features is 95.36%, with left stripes achieving a peak accuracy of 99.37%. These results highlight the model's excellent recognition capabilities. Our research provides a valuable and practical approach to the individual identification of rare and endangered animals, offering significant potential for improving conservation efforts. [ABSTRACT FROM AUTHOR]

Brain tumors are a leading cause of death globally, with numerous types varying in malignancy, and only 12% of adults diagnosed with brain cancer survive beyond five years. This research introduces a hyperparametric convolutional neural network (CNN) model to identify brain tumors, with significant practical implications. By fine-tuning the hyperparameters of the CNN model, we optimize feature extraction and systematically reduce model complexity, thereby enhancing the accuracy of brain tumor diagnosis. The critical hyperparameters include batch size, layer counts, learning rate, activation functions, pooling strategies, padding, and filter size. The hyperparameter-tuned CNN model was trained on three different brain MRI datasets available at Kaggle, producing outstanding performance scores, with an average value of 97% for accuracy, precision, recall, and F1-score. Our optimized model is effective, as demonstrated by our methodical comparisons with state-of-the-art approaches. Our hyperparameter modifications enhanced the model performance and strengthened its capacity for generalization, giving medical practitioners a more accurate and effective tool for making crucial judgments regarding brain tumor diagnosis. Our model is a significant step in the right direction toward trustworthy and accurate medical diagnosis, with practical implications for improving patient outcomes. [ABSTRACT FROM AUTHOR]

This study aimed to (1) replicate a deep-learning-based model for cerebral aneurysm segmentation in TOF-MRAs, (2) improve the approach by testing various fully automatic pre-processing pipelines, and (3) rigorously validate the model's transferability on independent, external test-datasets. A convolutional neural network was trained on 235 TOF-MRAs acquired on local scanners from a single vendor to segment intracranial aneurysms. Different pre-processing pipelines including bias field correction, resampling, cropping and intensity-normalization were compared regarding their effect on model performance. The models were tested on independent, external same-vendor and other-vendor test-datasets, each comprised of 70 TOF-MRAs, including patients with and without aneurysms. The best-performing model achieved excellent results on the external same-vendor test-dataset, surpassing the results of the previous publication with an improved sensitivity (0.97 vs. ~ 0.86), a higher Dice score coefficient (DSC, 0.60 ± 0.25 vs. 0.53 ± 0.31), and an improved false-positive rate (0.87 ± 1.35 vs. ~ 2.7 FPs/case). The model further showed excellent performance in the external other-vendor test-datasets (DSC 0.65 ± 0.26; sensitivity 0.92, 0.96 ± 2.38 FPs/case). Specificity was 0.38 and 0.53, respectively. Raising the voxel-size from 0.5 × 0.5×0.5 mm to 1 × 1×1 mm reduced the false-positive rate seven-fold. This study successfully replicated core principles of a previous approach for detecting and segmenting cerebral aneurysms in TOF-MRAs with a robust, fully automatable pre-processing pipeline. The model demonstrated robust transferability on two independent external datasets using TOF-MRAs from the same scanner vendor as the training dataset and from other vendors. These findings are very encouraging regarding the clinical application of such an approach. [ABSTRACT FROM AUTHOR]

Aim Methods Results Conclusions This study aimed to explore the value of ultrasound (US) images in chronic kidney disease (CKD) screening by constructing a CKD screening model based on grey‐scale US images.According to the CKD diagnostic criteria, 1049 patients from Tongde Hospital of Zhejiang Province were retrospectively enrolled in the study. A total of 4365 renal US images were collected from these patients. Convolutional neural networks were used for feature extractions and a screening model was constructed by fusing ResNet34 and texture features to identify CKD and its stage. A comparative analysis was performed to compare the diagnosis results of the model with physicians.When diagnosing CKD or non‐CKD, the receiver operating characteristic curve (AUC) of our model was 0.918 and that of the senior physician group was 0.869 (p < .05). For the diagnosis of CKD stage, the AUC of our model for CKD G1–G3 was 0.781, 0.880, and 0.905, respectively, while the AUC of the senior physician group for CKD G1–G3 was 0.506, 0.586, and 0.796, respectively; all differences were statistically significant (p < .05). The diagnostic efficiency of our model for CKD G4 and G5 reached the level of the senior physicians group. Specifically, the AUC of our model for CKD G4‐G5 was 0.867 and 0.931, respectively, while the AUC of the senior physician group for CKD G4‐G5 was 0.838 and 0.963, respectively (all p > .05).Our deep learning radiomics model is more effective than senior physicians in the diagnosis of early CKD. [ABSTRACT FROM AUTHOR]

Smart agriculture has shifted the paradigm by integrating advanced technologies, particularly weed management. This paper introduces an innovative approach to weed control by applying a Wavelet-based Convolution Neural Network (WCNN). In the era of precision agriculture, our study explores the integration of WCNN into real-world scenarios, emphasizing its adaptability to diverse environmental conditions. Utilizing the spatial-frequency analysis features of wavelets and convolutional neural networks, the WCNN model is the most effective at finding weeds, classifying them, and managing them specifically in agricultural fields in real-time. This research contributes to the scientific discourse on smart agriculture and addresses the challenges of invasive weeds, presenting a sustainable solution for optimizing resource utilization. Our investigation includes a detailed exploration of WCNN’s adaptive learning mechanisms and dynamic adjustment to changing agricultural landscapes. The model seamlessly integrates with existing smart farming infrastructure, showcasing a substantial reduction in manual intervention and a simultaneous increase in agricultural productivity. We incorporate fog computing and resource optimization into our framework, enhancing the efficiency of onboard data processing. To evaluate the real-world efficacy of WCNN, we conducted comprehensive experiments in texture classification and image labelling using two distinct datasets: the plant seedling and soybean weed datasets. Results demonstrate the superior performance of WCNN, achieving higher accuracy in training and test scenarios with significantly fewer parameters than traditional CNNs. For the soybean weed dataset, WCNN achieved remarkable accuracy in the training (0.9970) and testing (0.9987) phases, with correspondingly low losses of 0.0109 and 0.0048. The WCNN model demonstrated high accuracy during training (0.9739) and testing (0.9902), with minimal losses of 0.0898 and 0.0239 in the plant seedling dataset. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Landslide disasters, due to their widespread distribution and clustered occurrences, pose a significant threat to human society. Rainfall is considered a primary triggering factor, and the frequent clustering of landslides underscores the importance of early warning systems for regional landslide disasters in preventing and mitigating rainfall-induced landslides. Research on early warning models is crucial for accurately predicting rainfall-induced landslides. However, traditional models face challenges such as the complexity of landslide causes, insufficient data, and limited analysis methods, resulting in low accuracy and inadequate precision. This study focuses on Fujian Province, China, proposing a four-step process for building a regional landslide early warning model based on machine learning. The process includes data integration and cleaning, sample set construction, model training and validation, and practical application. By integrating and cleaning the latest and most detailed data, a training sample set (15,589 samples) for the regional landslide disaster early warning model is established. Three machine learning algorithms--Random Forest, Multilayer Perceptron, and Convolutional Neural Network--are employed and compared, the evaluation results indicated that the RF-based warning model achieved an accuracy of 0.930-0.957 and an AUC value of 0.955. The CNNbased warning model demonstrated an accuracy of 0.945-0.948 with an AUC value of 0.940. The MLP-based warning model achieved an accuracy of 0.930-0.953 and an AUC value of 0.930. The results showed comparable accuracy metrics among the three models, with RF exhibiting a significant advantage in AUC values. Finally, the models are applied to the regional landslide disasters induced by heavy rainfall in Fujian Province on 5 August 2021. The results showed that in the binary classification warning strategy, the accuracy of the Random Forest and Convolutional Neural Network was 92.9%, while that of the Multilayer Perceptron was 85.8%, all performing well. In the multi-classification hierarchical warning strategy, the Random Forest excelled, while the performance of the Convolutional Neural Network and Multilayer Perceptron was relatively limited. The findings of this study Frontiers in Earth Science 01 frontiersin.org Liu et al. 10.3389/feart.2024.1419421 contribute to valuable attempts in landslide disaster warning model research, with anticipated further improvements through the gradual accumulation of samples and practical application verification. [ABSTRACT FROM AUTHOR]

In recent years, graph-based learning methods have gained significant traction in point-of-interest (POI) recommendation systems due to their strong generalization capabilities. These approaches commonly transform user check-in records into graph-structured data and leverage graph neural networks (GNNs) to model the representations of both POIs and users. Despite their effectiveness, GNNs face inherent limitations in message passing, which can impede the deep extraction of meaningful representations from the graph structure. To mitigate this challenge, we introduce a novel framework, Multi-view Contextual Graphs via Convolutional Neural Networks for Point-of-Interest Recommendation (MCGRec). The MCGRec framework consists of three primary components. Firstly, it employs a personalized PageRank (PPR) sampling technique based on super nodes to transform the graph-structured data into a grid-like feature matrix. This step is crucial as it prepares the data for subsequent processing by convolutional neural networks (CNNs), which are adept at extracting spatial features from grid-like structures. Secondly, a CNN is utilized to extract the representations of POIs from the constructed feature matrix. The usage of CNNs enables the capture of local patterns and hierarchical features within the data, which are essential for accurate POI representation. Lastly, MCGRec incorporates a novel approach for estimating user preferences that integrates both geographical and temporal factors, thereby providing a more comprehensive model of users' behaviors. To evaluate the effectiveness of our proposed method, we conduct extensive experiments on real-world datasets. Our results demonstrate that MCGRec outperforms state-of-the-art POI recommendation methods, showcasing its superiority in terms of recommendation accuracy. [ABSTRACT FROM AUTHOR]

In sports medicine, anterior cruciate ligament (ACL) injuries are common and have a major effect on knee joint stability. For the sake of prognosis evaluation and treatment planning, an accurate clinical auxiliary diagnosis of ACL injuries is essential. Although existing deep learning techniques for ACL diagnosis work well on single datasets, research on cross-domain data transfer is still lacking. Building strong domain-adaptive diagnostic models requires addressing domain disparities in ACL magnetic resonance imaging (MRI) from different hospitals and making efficient use of multiple ACL datasets. This work uses the publicly available KneeMRI dataset from Croatian hospitals coupled with the publicly available MRnet dataset from Stanford University to investigate domain adaptation and transfer learning models. First, an optimized model efficiently screens training data in the source domain to find unusually misclassified occurrences. Subsequently, before being integrated into the contrastive learning module, a target domain feature extraction module processes features of target domain samples to improve extraction efficiency. By using contrastive learning between positive and negative sample pairs from source and target domains, this method makes domain adaptation easier and improves the efficacy of ACL auxiliary diagnostic models. Utilizing a spatially augmented ResNet-18 backbone network, the suggested approach produces notable enhancements in experimentation. To be more precise, the AUC for transfer learning improved by 3.5% from MRnet to KneeMRI and by 2.5% from KneeMRI to MRnet (from 0.845 to 0.870). This method shows how domain transfer can be used to improve diagnostic accuracy on a variety of datasets and effectively progresses the training of a strong ACL auxiliary diagnostic model. [ABSTRACT FROM AUTHOR]

Rotated object detection is a crucial task in aerial image analysis. To address challenges such as multi-directional object rotation, complex backgrounds with occlusions, and the trade-off between speed and accuracy in remote sensing images, this paper introduces a lightweight progressive fusion calibration network for rotated object detection (LPFC-RDet). The network comprises three main modules: the Retentive Meet Transformers (RMT) feature extraction block, the Progressive Fusion Calibration module (PFC), and the Shared Group Convolution Lightweight detection head (SGCL). The RMT feature extraction block integrates a retentive mechanism with global context modeling to learn rotation-insensitive features. The PFC module employs pixel-level, local-level, and global-level weights to calibrate features, enhancing feature extraction from occluded objects while suppressing background interference. The SGCL detection head uses decoupled detection tasks and shared group convolution layers to achieve parameter sharing and feature interaction, improving accuracy while maintaining a lightweight structure. Experimental results demonstrate that our method surpasses state-of-the-art detectors on three widely used remote sensing object datasets: HRSC2016, UCAS_AOD, and DOTA. [ABSTRACT FROM AUTHOR]

The accurate identification of low-voltage distribution substation topology plays a crucial role in research applications such as line loss management, fault location, imbalance correction, and intelligent operation and maintenance of substations. In response to the challenges posed by the large scale of users and the complex connection relationships in low-voltage distribution substations, which complicate the identification of their topology, a method for identifying low-voltage distribution substation topology based on user profiling technology is proposed. This method is supported by big data technology and introduces the concept of user profiling into the research of substation topology identification. Firstly, by deeply studying the theoretical knowledge of the relationship between the supply of distribution transformers and the power consumption of substation users, as well as the voltage similarity, we establish the feature labels for electricity coefficient and voltage similarity. Then, we use the continuous relaxation method and branch-and-bound method to solve the electricity coefficient matrix; and use the Gaussian kernel function to solve the voltage similarity matrix. Finally, by constructing a comprehensive attribution matrix using the electricity coefficient matrix and the voltage similarity matrix, employing a convolutional neural network to cluster and solve the comprehensive attribution matrix, outputting user-area attribution information, forming user-area attribution profiles, and completing the identification of area attribution relationships in low-voltage distribution systems. Simulation results demonstrate that the proposed method not only effectively identifies the area information to which low-voltage users belong but also discerns the connection relationships between users and area transformers. [ABSTRACT FROM AUTHOR]

Background and objective: Epilepsy, which is associated with neuronal damage and functional decline, typically presents patients with numerous challenges in their daily lives. An early diagnosis plays a crucial role in managing the condition and alleviating the patients' suffering. Electroencephalogram (EEG)- based approaches are commonly employed for diagnosing epilepsy due to their effectiveness and non-invasiveness. In this study, a classification method is proposed that use fast Fourier Transform (FFT) extraction in conjunction with convolutional neural networks (CNN) and long short-term memory (LSTM) models. Methods: Most methods use traditional frameworks to classify epilepsy, we propose a new approach to this problem by extracting features from the source data and then feeding them into a network for training and recognition. It preprocesses the source data into training and validation data and then uses CNN and LSTM to classify the style of the data. Results: Upon analyzing a public test dataset, the top-performing features in the fully CNN nested LSTM model for epilepsy classification are FFT features among three types of features. Notably, all conducted experiments yielded high accuracy rates, with values exceeding 96% for accuracy, 93% for sensitivity, and 96% for specificity. These results are further benchmarked against current methodologies, showcasing consistent and robust performance across all trials. Our approach consistently achieves an accuracy rate surpassing 97.00%, with values ranging from 97.95 to 99.83% in individual experiments. Particularly noteworthy is the superior accuracy of our method in the AB versus (vs.) CDE comparison, registering at 99.06%. Conclusion: Our method exhibits precise classification abilities distinguishing between epileptic and non-epileptic individuals, irrespective of whether the participant's eyes are closed or open. Furthermore, our technique shows remarkable performance in effectively categorizing epilepsy type, distinguishing between epileptic ictal and interictal states versus non-epileptic conditions. An inherent advantage of our automated classification approach is its capability to disregard EEG data acquired during states of eye closure or eye-opening. Such innovation holds promise for real-world applications, potentially aiding medical professionals in diagnosing epilepsy more efficiently. [ABSTRACT FROM AUTHOR]

Head pose estimation methods can be generally classified into two categories: model-based and appearance-based methods. The model-based approach relies on facial landmarks for three-dimensional reconstruction, aiming to achieve high-precision results. However, this method is heavily dependent on the accuracy of these landmarks. The appearance-based approach utilizes images as input and employs feature extraction and calculations to generate outcomes. While the appearance-based method boasts greater robustness, its accuracy falls short of the former. In this paper, a new and effective hybrid method is proposed. This hybrid approach combines the strengths of both methods. Unlike the conventional model-based methods, the proposed method regards the facial landmarks in 2D images as a sequence of neural network inputs and then obtains the head pose estimation results for users by neural network regression. The proposed method solves the fuzzy rotation labeling problem by using a rotation matrix representation, introducing a 6D rotation matrix representation as an intermediate state of the rotation matrix to achieve effective direct regression. Introducing face processing enhances the robustness of the model in cross-dataset scenarios. The proposed method achieves remarkable results based on imprecise face recognition and a simplistic model. The proposed method can be divided into three parts. First, the proposed method applies face processing on the input image; second, the method detects facial landmarks; and third, it converts these facial landmarks into sequences and obtains the 6D rotation representation of the head pose by regression. Extensive experiments on the publicly available BIWI, PRIMA, and DrivFace datasets show that this method is functional and performs better than other state-of-the-art methods. Compared to other methods, this approach demonstrates an average performance improvement of at least 10% across the dataset. [ABSTRACT FROM AUTHOR]

This study aimed to investigate the performance and factors affecting the species classification of convolutional neural network (CNN) architecture using whole-part and earlywood-part cross-sectional datasets of six Korean Quercus species. The accuracy of species classification for each condition was analyzed using the datasets, data augmentation, and optimizers—stochastic gradient descent (SGD), adaptive moment estimation (Adam), and root mean square propagation (RMSProp)—based on a CNN architecture with three to four convolutional layers. The model trained with the augmented dataset yielded significantly superior results in terms of classification accuracy compared to the model trained with the non-augmented dataset. The augmented dataset was the only factor affecting classification accuracy in the final five epochs. In contrast, four factors in the entire epoch, such as the Adam and SGD optimizers and the earlywood-part and whole-part datasets, affected species classification. The arrangement of earlywood vessels, broad rays, and axial parenchyma was identified as a major influential factor in the CNN species classification using gradient-weighted class activation mapping (Grad-CAM) analysis. The augmented whole-part dataset with the Adam optimizer achieved the highest classification accuracy of 85.7% during the final five epochs of the test phase. [ABSTRACT FROM AUTHOR]

Leukemia is a type of blood cell cancer that is in the bone marrow's blood-forming cells. Two types of Leukemia are acute and chronic; acute enhances fast and chronic growth gradually which are further classified into lymphocytic and myeloid leukemias. This work evaluates a unique deep convolutional neural network (CNN) classifier that improves identification precision by carefully examining concatenated peptide patterns. The study uses leukemia protein expression for experiments supporting two different techniques including independence and applied cross-validation. In addition to CNN, multilayer perceptron (MLP), gated recurrent unit (GRU), and recurrent neural network (RNN) are applied. The experimental results show that the CNN model surpasses competitors with its outstanding predictability in independent and cross-validation testing applied on different features extracted from protein expressions such as amino acid composition (AAC) with a group of AAC (GAAC), tripeptide composition (TPC) with a group of TPC (GTPC), and dipeptide composition (DPC) for calculating its accuracies with their receiver operating characteristic (ROC) curve. In independence testing, a feature expression of AAC and a group of GAAC are applied using MLP and CNN modules, and ROC curves are achieved with overall 100% accuracy for the detection of protein patterns. In cross-validation testing, a feature expression on a group of AAC and GAAC patterns achieved 98.33% accuracy which is the highest for the CNN module. Furthermore, ROC curves show a 0.965% extraordinary result for the GRU module. The findings show that the CNN model is excellent at figuring out leukemia illnesses from protein expressions with higher accuracy. [ABSTRACT FROM AUTHOR]

In recent years, cloud computing applications have facilitated the distribution of heterogeneous, unstructured digitized data among users' social networks of varying opinions. Text processing at a large scale requires high-precision computational techniques, which increases the computational burden. The advent of big data analytics along with Natural Language Processing is a powerful factor in improving the efficiency of processing large-scale text data, using the kernel of the MapReduce big data analytics frameworks that allow parallelization of large computational operations. In this paper, we propose an intelligent sentiment prediction approach based on deep learning, batch, and streaming big data analytics. In fact, our main objective is to take advantages of the powerful tools provided by the distributed platforms, such as, Hadoop and Spark to preprocess streaming data. This involves various tasks such as cleaning the data, reducing its size, minimizing access time, and decreasing storage volume. This step prepares the big streaming data to be fully exploited by Deep Learning models. This work includes a research study on processing big data related to short volume scripts based on batch and streaming distributed frameworks as well as deep learning approaches in Natural Language Processing. We detail our idea for analyzing short texts to determine their semantic context and categorize them into pros and cons poles. There were different stages in building this model, the first involving data reduction and refinement using selected features and big data analysis tools. In the second stage, words are embedded by global vector to be computed in layers of convolutional and recurrent neural networks. The experimental study and the analysis of the results confirm the usefulness of our proposed model and its superiority over the main approaches studied in the literature. Our model achieved a performance of 96% in terms of accuracy. [ABSTRACT FROM AUTHOR]

Objectives: To perform a systematic review on artificial intelligence (AI) performances to detect urinary stones. Methods: A PROSPERO-registered (CRD473152) systematic search of Scopus, Web of Science, Embase, and PubMed databases was performed to identify original research articles pertaining to AI stone detection or measurement, using search terms ("automatic" OR "machine learning" OR "convolutional neural network" OR "artificial intelligence" OR "detection" AND "stone volume"). Risk-of-bias (RoB) assessment was performed according to the Cochrane RoB tool, the Joanna Briggs Institute Checklist for nonrandomized studies, and the Checklist for Artificial Intelligence in Medical Imaging (CLAIM). Results: Twelve studies were selected for the final review, including three multicenter and nine single-center retrospective studies. Eleven studies completed at least 50% of the CLAIM checkpoints and only one presented a high RoB. All included studies aimed to detect kidney (5/12, 42%), ureter (2/12, 16%), or urinary (5/12, 42%) stones on noncontrast computed tomography (NCCT), but 42% intended to automate measurement. Stone distinction from vascular calcification interested two studies. All studies used AI machine learning network training and internal validation, but a single one provided an external validation. Trained networks achieved stone detection, with sensitivity, specificity, and accuracy rates ranging from 58.7% to 100%, 68.5% to 100%, and 63% to 99.95%, respectively. Detection Dice score ranged from 83% to 97%. A high correlation between manual and automated stone volume (r = 0.95) was noted. Differentiate distal ureteral stones and phleboliths seemed feasible. Conclusions: AI processes can achieve automated urinary stone detection from NCCT. Further studies should provide urinary stone detection coupled with phlebolith distinction and an external validation, and include anatomical abnormalities and urologic foreign bodies (ureteral stent and nephrostomy tubes) cases. [ABSTRACT FROM AUTHOR]

Background: Legumes, such as lentils, field peas, Faba beans and chickpeas, are high in vitamins, fiber, important minerals and protein and can help avoid obesity and cardiovascular illnesses. They also contribute to ecosystem services, such as nitrogen fixation and resilience to environmental stresses. Despite a 60% increase in global pulse production from 2000 to 2021, a demand-supply gap, especially in South Asia, raises concerns about nutritional access. Since illnesses are currently an issue to the food security of faba beans, machine learning is required for efficient disease identification. Methods: This research employs Convolutional Neural Networks (CNNs) for robust Faba bean leaf disease identification. The CNN model is trained with diverse images representing specific diseases. The study focuses on diseases like Chocolate Spot, Faba Bean Gall, Rust and Healthy leaves. Image processing involves resizing, grayscale conversion and labeling. The CNN architecture includes eight convolutional layers, four max-pooling layers and three dropout layers. The model is trained using 80% of the dataset, validated using 20% and tested for accuracy. Result: The CNN model achieves an accuracy of 99.37% during training and 89.69% during validation after 75 epochs. Confusion matrix and classification report illustrate the model’s performance. It shows high precision, recall and F1 scores for each class, indicating balanced performance. Chocolate Spot and Rust exhibit the highest precision and F1 scores. The overall accuracy is 91%, comparable to other studies on Faba bean disease detection. The study presents a CNN-based disease identification system for Faba beans, demonstrating high accuracy and balanced performance across different diseases. The model’s effectiveness is comparable to other advanced techniques. The research highlights the potential of machine learning in optimizing disease management for Faba beans. Future work could explore a broader range of diseases and incorporate hybrid machine learning algorithms for further improvement. [ABSTRACT FROM AUTHOR]

Rolling bearing fault diagnosis based on convolutional neural network is greatly effective for bearing maintenance, and it is of great significance for ensuring the safe operation of rotating machinery. However, the traditional convolutional neural network models only focus on the single-scale feature and ignore the multi-scale deep information, which results in low performance for conducting the complex fault diagnosis problem. Aiming at this problem, we propose an improved fault diagnosis method of rolling bearings based on multi-scale attention convolutional neural network. In the first stage, a one-dimensional convolutional neural network model integrating feature extraction and fault intelligence classification is established. Subsequently, a multi-scale structure with serial layer skipping connection is constructed to extract multi-scale features in different reception fields, and the "Concat" operator is used to fuse the extracted features of each layer with SE attention mechanism in order to obtain more important feature information. And then, the fault samples are input into the network model to realize the end-to-end fault diagnosis of rolling bearings result from the nonlinear fitting ability of deep learning. Extensive experiments on two well-known rolling bearing datasets validate that the proposed method not only achieves higher fault diagnosis accuracy on the fault data sets under constant load conditions, but also exhibits strong fault identification and transfer diagnosis ability on the fault data sets under variable load conditions. [ABSTRACT FROM AUTHOR]

Underwater segmentation technology achieves effective monitoring of fish biological information through accurate identification of fish species and precise estimation of their quantities. It serves as an effective approach to enhance the informatization level of aquaculture and promote intelligent management in fisheries. However, the complex and diverse underwater environment, coupled with poor visibility, results in blurry and lower-quality underwater fish images. The current image segmentation methods, when applied to fish segmentation, exhibit low accuracy and inadequate generalization capabilities. This paper proposes an image segmentation model based on Atrous Spatial Pyramid Pooling (ASPP) to address these challenges. The model aims to improve fish feature extraction and enhance the segmentation precision of fish images in complex underwater environments, thereby enhancing the accuracy and generalization capabilities of existing fish segmentation models. First, a multiscale feature extraction module (triangular atrous spatial pyramid multifeature fusion) based on dilated convolutional spatial pyramid pooling, which enhances the extraction of high-level semantic features of images through the triangular combination of multilayer dilated convolutional pyramid pooling and the adaptive channel attention module, is proposed. Second, a spatial attention module based on strip pooling (atrous strip pooling) is proposed, which further expands the receptive field of the attention mechanism by combining different expansion rates, enhances the correlations between pixels, and effectively captures spatial information. Finally, a decoder module based on multilayer semantic feature fusion is proposed. Through the processing and fusion of medium-, low-, and high-level semantic features, the model understands image content and performs accurate pixel-level segmentation. The proposed model is evaluated using a VOC-compliant dataset created from underwater fish images and validated against public and specific underwater fish datasets. The results demonstrate the successful application of the feature extraction and feature fusion modules in underwater fish image segmentation, achieving an average Intersection over Union (MIoU) of 85.49% in segmentation tasks. Compared to conventional segmentation models, the proposed model shows significant improvements, with an average increase of 3.8% in MIoU and 2.5% in balanced F-score (F1-score). The accurate segmentation of underwater fish images and extraction of vital biological information by this model provide a solid foundation for intelligent monitoring, including fish length measurement, weight estimation, and analysis of growth and health status. Moreover, the model offers a scientific framework for decision-making in aquaculture, driving advancements in precision and intelligent management practices in the field. [ABSTRACT FROM AUTHOR]

Existing text classification models insufficiently utilize global and local information in texts, leading to subpar classification performance, In response to this issue, a text classification model called global and local features dynamic fusion (GLFDF) is proposed. The GLFDF model was initially designed with a dynamic fusion enhancement module to dynamically control the inte- gration of global temporal features and local semantic features into specific positions of the word embedding matrix. Subsequently, the embedding matrix where global and local features fused was fed into a feature extraction module for further processing. Finally, the proposed model was tested on two public datasets, Ohsumed and THUCNews. Experimental results show that the GLFDF model achieves F1 scores of 63,24% and 92.50% on these datasets, respectively, surpassing other advanced text classification models and enhancing text classification performance. From the analy- sis of the ablation experiment, the dynamic fusion enhancement module fully makes the global temporal features and local semantic features of the text fused together, effectively solving the problem of insufficient use of global and local information in the text classification model. [ABSTRACT FROM AUTHOR]

On the Internet, online microblogging and social networks have experienced tremendous growth. Millions of individuals share opinions on social networking platforms including Twitter, Facebook, YouTube, and Microblogging sites. Millions of individuals express their thoughts on social platforms namely Twitter, Facebook, YouTube, and microblogging sites. As user‐generated content is growing to a huge extent on the internet, the analysis, extraction identification and classification of opinion and sentiment polarity of different aspects has now become a challenging issue in sentiment analysis. In sentiment analysis, aspect level opinion mining, opinion spam review detection, extraction and representation of most relevant accessed data from the user‐generated data is a difficult process. The word vector representation using existing deep learning techniques affects the performance in aspect level opinion mining. Spammers propagate false opinion for payment exchange, which leads to degradation of the financial growth, business, and fame of the organization. To overcome the limitations aspect level opinion mining is used to recognize sentiment and polarity of aspects in the specified text. This work presents various input feature vectors of deep learning approaches aiming to improve performance in aspect level opinion mining. Next, three meta‐heuristic algorithms with k‐means clustering approach for classification of opinion spam reviews on social media transit tweets is presented. The proposed approaches are evaluated and compared with other existing approaches as well as other benchmark datasets namely Restaurant and Amazon reviews. [ABSTRACT FROM AUTHOR]

With this work, we hypothesize that semantically enriching a user's text corpus using backtranslation and expansion modules can improve performance for author profiling tasks. To perform this textual enrichment, we translate an author's representative text. Translations are made from one language—the source language—into another—the target language—and then back to the original one. Finally, we expand an author's text by integrating the original version with the back‐translated one. Our framework includes these backtranslation and expansion modules followed by a SOTA classifier successfully employed for text classification. The framework is tested on three author profiling datasets from the last three years' shared tasks on fake news, hate speech, irony and stereotypes detection hosted at the CLEF conference for the PAN Lab. This work is an extension of our previous one where we just presented our main idea. Here we improve our framework, and we also investigate more languages and more datasets. Finally, a qualitative analysis is provided. The results confirm that the backtranslation and expansion add‐on modules improve model performance on all three datasets evaluated. [ABSTRACT FROM AUTHOR]

Accurate and reliable medium- and long-term load forecasting is crucial for the rational planning and operation of power systems. However, existing methods often struggle to accurately extract and capture long-term dependencies in load data, leading to poor predictive accuracy. Therefore, this paper proposes a medium- and long-term residential load forecasting method based on FEDformer, aiming to capture long-term temporal dependencies of load data in the frequency domain while considering factors such as electricity prices and temperature, ultimately improving the accuracy of medium- and long-term load forecasting. The proposed model employs Discrete Cosine Transform (DCT) for frequency domain transformation of time-series data to address the Gibbs phenomenon caused by the use of Discrete Fourier Transform (DFT) in FEDformer. Additionally, causal convolution and attention mechanisms are applied in the frequency domain to enhance the model's capability to capture long-term dependencies. The model is evaluated using real-world load data from power systems, and experimental results demonstrate that the proposed model effectively learns the temporal and nonlinear characteristics of load data. Compared to other baseline models, DCTformer improves prediction accuracy by 37.5% in terms of MSE, 26.9% in terms of MAE, and 26.24% in terms of RMSE. [ABSTRACT FROM AUTHOR]

Low-light images are prevalent in intelligent monitoring and many other applications, with low brightness hindering further processing. Although low-light image enhancement can reduce the influence of such problems, current methods often involve a complex network structure or many iterations, which are not conducive to their efficiency. This paper proposes a Zero-Reference Camera Response Network using a camera response model to achieve efficient enhancement for arbitrary low-light images. A double-layer parameter-generating network with a streamlined structure is established to extract the exposure ratio K from the radiation map, which is obtained by inverting the input through a camera response function. Then, K is used as the parameter of a brightness transformation function for one transformation on the low-light image to realize enhancement. In addition, a contrast-preserving brightness loss and an edge-preserving smoothness loss are designed without the requirement for references from the dataset. Both can further retain some key information in the inputs to improve precision. The enhancement is simplified and can reach more than twice the speed of similar methods. Extensive experiments on several LLIE datasets and the DARK FACE face detection dataset fully demonstrate our method's advantages, both subjectively and objectively. [ABSTRACT FROM AUTHOR]

In this paper, artificial intelligence (AI) technology is applied to the electromagnetic imaging of anisotropic objects. Advances in magnetic anomaly sensing systems and electromagnetic imaging use electromagnetic principles to detect and characterize subsurface or hidden objects. We use measured multifrequency scattered fields to calculate the initial dielectric constant distribution of anisotropic objects through the backpropagation scheme (BPS). Later, the estimated multifrequency permittivity distribution is input to a convolutional neural network (CNN) for the adaptive moment estimation (ADAM) method to reconstruct a more accurate image. In the meantime, we also improve the definition of loss function in the CNN. Numerical results show that the improved loss function unifying the structural similarity index measure (SSIM) and root mean square error (RMSE) can effectively enhance image quality. In our simulation environment, noise interference is considered for both TE (transverse electric) and TM (transverse magnetic) waves to reconstruct anisotropic scatterers. Lastly, we conclude that multifrequency reconstructions are more stable and precise than single-frequency reconstructions. [ABSTRACT FROM AUTHOR]

Dunes are the primary geomorphological type in deserts, and the distribution of dune morphologies is of significant importance for studying regional characteristics, formation mechanisms, and evolutionary processes. Traditional dune morphology classification methods rely on visual interpretation by humans, which is not only time-consuming and inefficient but also subjective in classification judgment. These issues have impeded the intelligent development of dune morphology classification. However, convolutional neural network (CNN) models exhibit robust feature representation capabilities for images and have achieved excellent results in image classification, providing a new method for studying dune morphology classification. Therefore, this paper summarizes five typical dune morphologies in the deserts of western Inner Mongolia, which can be used to define and describe most of the dune types in Chinese deserts. Subsequently, field surveys and the experimental collection of unmanned aerial vehicle (UAV) orthoimages for different dune types were conducted. Five different types of dune morphology datasets were constructed through manual segmentation, automatic rule segmentation, random screening, and data augmentation. Finally, the classification of dune morphologies and the exploration of dataset construction methods were conducted using the VGG16 and VGG19 CNN models. The classification results of dune morphologies were comprehensively analyzed using different evaluation metrics. The experimental results indicate that when the regular segmentation scale of UAV orthoimages is 1024 × 1024 pixels with an overlap of 100 pixels, the classification accuracy, precision, recall, and F1-Score of the VGG16 model reached 97.05%, 96.91%, 96.76%, and 96.82%, respectively. The method for constructing a dune morphology dataset from automatically segmented UAV orthoimages provides a reference value for the study of large-scale dune morphology classification. [ABSTRACT FROM AUTHOR]

The vulnerability of modern neural networks to random noise and deliberate attacks has raised concerns about their robustness, particularly as they are increasingly utilized in safety- and security-critical applications. Although recent research efforts were made to enhance robustness through retraining with adversarial examples or employing data augmentation techniques, a comprehensive investigation into the effects of training data perturbations on model robustness remains lacking. This paper presents the first extensive empirical study investigating the influence of data perturbations during model retraining. The experimental analysis focuses on both random and adversarial robustness, following established practices in the field of robustness analysis. Various types of perturbations in different aspects of the dataset are explored, including input, label, and sampling distribution. Single-factor and multi-factor experiments are conducted to assess individual perturbations and their combinations. The findings provide insights into constructing high-quality training datasets for optimizing robustness and recommend the appropriate degree of training set perturbations that balance robustness and correctness, and contribute to understanding model robustness in deep learning and offer practical guidance for enhancing model performance through perturbed retraining, promoting the development of more reliable and trustworthy deep learning systems for safety-critical applications. [ABSTRACT FROM AUTHOR]

Rapid and accurate diagnostic tests are fundamental for improving patient outcomes and combating infectious diseases. The Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) Cas12a‐based detection system has emerged as a promising solution for on‐site nucleic acid testing. Nonetheless, the effective design of CRISPR RNA (crRNA) for Cas12a‐based detection remains challenging and time‐consuming. In this study, we propose an enhanced crRNA design system with deep learning for Cas12a‐mediated diagnostics, referred to as EasyDesign. This system employs an optimized convolutional neural network (CNN) prediction model, trained on a comprehensive data set comprising 11,496 experimentally validated Cas12a‐based detection cases, encompassing a wide spectrum of prevalent pathogens, achieving Spearman's ρ = 0.812. We further assessed the model performance in crRNA design for four pathogens not included in the training data: Monkeypox Virus, Enterovirus 71, Coxsackievirus A16, and Listeria monocytogenes. The results demonstrated superior prediction performance compared to the traditional experiment screening. Furthermore, we have developed an interactive web server (https://crispr.zhejianglab.com/) that integrates EasyDesign with recombinase polymerase amplification (RPA) primer design, enhancing user accessibility. Through this web‐based platform, we successfully designed optimal Cas12a crRNAs for six human papillomavirus (HPV) subtypes. Remarkably, all the top five predicted crRNAs for each HPV subtype exhibited robust fluorescent signals in CRISPR assays, thereby suggesting that the platform could effectively facilitate clinical sample testing. In conclusion, EasyDesign offers a rapid and reliable solution for crRNA design in Cas12a‐based detection, which could serve as a valuable tool for clinical diagnostics and research applications. Highlights: Advanced crRNA design system: Developed EasyDesign, utilizing a convolutional neural network (CNN) trained with over 11,000 diagnostic‐target pairs to enable the creation of highly sensitive crRNAs for Cas12a‐based nucleic acid diagnostics.Proven predictive capabilities: EasyDesign demonstrates superior predictive performance for crRNA‐mediated Cas12a detection, validated through its successful application in designing diagnostics for a variety of viruses in clinical settings.User‐friendly web platform: Features an intuitive web platform that streamlines the creation of CRISPR/Cas12a‐based diagnostic tools, thereby enhancing accessibility and usability for both researchers and clinicians. [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]

Numerical simulation of multiphase flow in porous media is an important step in understanding the dynamic behavior of geologic CO 2 storage (GCS). Scaling up GCS requires fast and accurate high-resolution modeling of the storage reservoir pressure and saturation plume migration; however, such modeling is challenging due to the high computational costs of traditional physics-based simulations. Deep learning models trained with numerical simulation data can provide a fast and reliable alternative to expensive physics-based numerical simulations. We propose a Stochastic pix2vid neural network architecture for solving multiphase fluid flow problems with significant speed, accuracy, and efficiency. The Stochastic pix2vid model is designed based on the principles of computer vision and video synthesis and is able to generate dynamic spatiotemporal predictions of fluid flow from static reservoir models, closely mimicking the performance of traditional numerical simulation. We apply the Stochastic pix2vid model to a highly-complex CO 2 -water multiphase problem with a wide range of reservoir models in terms of porosity and permeability heterogeneity, facies distribution, and injection configurations. The Stochastic pix2vid method is first-of-its-kind in static-to-dynamic prediction of reservoir behavior, where a single static input is mapped to its dynamic response with a fixed number of timesteps. The Stochastic pix2vid method provides notable performance in highly heterogeneous geologic formations and complex estimation such as CO 2 saturation and pressure buildup plume determination. The trained model can serve as a general-purpose, static-to-dynamic (image-to-video) alternative to traditional numerical reservoir simulation of 2D CO 2 injection problems with up to 6,500 × speedup compared to traditional numerical simulation using the MATLAB Reservoir Simulation Toolbox. [ABSTRACT FROM AUTHOR]

Deep brain stimulation (DBS) of the subthalamic nucleus (STN) is an effective therapy in ameliorating the motor symptoms of Parkinson disease. However, postoperative optimal contact selection is crucial for achieving the best outcome of deep brain stimulation of the subthalamic nucleus surgery, but the process is currently a trial-and-error and time-consuming procedure that relies heavily on surgeons' clinical experience. In this study, we propose a structural brain connectivity guided optimal contact selection method for deep brain stimulation of the subthalamic nucleus. Firstly, we reconstruct the DBS electrode location and estimate the stimulation range using volume of tissue activated from each DBS contact. Then, we extract the structural connectivity features by concatenating fractional anisotropy and the number of streamlines features of activated regions and the whole brain regions. Finally, we use a convolutional neural network with convolutional block attention module to identify the structural connectivity features for the optimal contact selection. We review the data of 800 contacts from 100 patients with Parkinson disease for the experiment. The proposed method achieves promising results, with the average accuracy of 97.63%, average precision of 94.50%, average recall of 94.46%, and average specificity of 98.18%, respectively. Our method can provide the suggestion for optimal contact selection. Our proposed method can improve the efficiency and accuracy of DBS optimal contact selection, reduce the dependence on surgeons' experience, and has the potential to facilitate the development of advanced DBS technology. [ABSTRACT FROM AUTHOR]

Remote sensing (RS) images play an indispensable role in many key fields such as environmental monitoring, precision agriculture, and urban resource management. Traditional deep convolutional neural networks have the problem of limited receptive fields. To address this problem, this paper introduces a hybrid network model that combines the advantages of CNN and Transformer, called MBT-UNet. First, a multi-branch encoder design based on the pyramid vision transformer (PVT) is proposed to effectively capture multi-scale feature information; second, an efficient feature fusion module (FFM) is proposed to optimize the collaboration and integration of features at different scales; finally, in the decoder stage, a multi-scale upsampling module (MSUM) is proposed to further refine the segmentation results and enhance segmentation accuracy. We conduct experiments on the ISPRS Vaihingen dataset, the Potsdam dataset, the LoveDA dataset, and the UAVid dataset. Experimental results show that MBT-UNet surpasses state-of-the-art algorithms in key performance indicators, confirming its superior performance in high-precision remote sensing image segmentation tasks. [ABSTRACT FROM AUTHOR]

Concrete cracking is one of the most significant damage types in reinforced concrete structures due to its potential to adversely affect durability, safety, and serviceability and even reduce the bearing capacity during operation. Thus, damage inspection of damage caused by concrete cracking is important for management, maintenance, and structural assessment for both damaged and undamaged existing buildings but with concrete cracking after a long time of use that needs reconstruction or renovation. This study provides an improved building damage inspection approach by applying Unmanned Aerial Vehicles (UAVs) and state‐of‐the‐art deep learning algorithms to detect concrete cracks on building surfaces. Two distinct architectures for Convolutional Neural Networks (CNNs), namely ResNet50 and YOLOv8 based on classification, and object detection approaches to create a total of 11 models are established, trained, and compared. The classification models attained accuracy levels exceeding 99%, whereas the object detection models achieved approximately 85%. All models effectively identified and accurately located concrete cracks on building surfaces. Besides, the CNN models' capacity to detect cracks is influenced by a variety of model hyperparameters, encompassing factors such as model architecture, the number of network layers, different training dataset sizes, and the quantity of trainable parameters necessary to learn the specific features of detection targets during the training process. The results of this study ultimately demonstrate that the proposed approach can yield accurate detection results and holds high potential for application in crack inspection to advance automatic damage inspection in building structures to a greater extent. In addition, it is important to note that a universal rule cannot be established rule as a larger and more complex model, or an increased number of trainable parameters, necessarily leads to improved detection performance. Models that are trained from scratch using local datasets might not necessarily result in enhanced performance in comparison to the improvements gained through fine‐tuning via transfer learning. Therefore, an appropriate training type, dataset size, task complexity, computational resources, and time demands to achieve a balance between accuracy and efficiency should be considered for specific application scenarios. [ABSTRACT FROM AUTHOR]