Search Results (737)

The transition toward sustainable power systems requires load forecasting methods that can support renewable integration under increasing uncertainty. However, many deep learning models mix historical load, temporal priors, and external drivers in black-box structures, and often assume that true future driver values are available. To address these issues, this study proposes ADDF (Automatic Driver Discovery and Fusion), a semi-explicit self-driven framework for multi-step load interval forecasting. ADDF organizes historical load, calendar priors, and external drivers into three functional branches to distinguish load inertia, temporal regularity, and external forcing. The Driver Branch estimates future driver states under practical information constraints and uses dynamic gating to screen useful driving information. The three branch representations are adaptively integrated through Three-Way Fusion, followed by bounded residual correction to generate multi-step quantile forecasts. Experiments on the Panama electricity load dataset and ETTh1 dataset under one-step and 24-step settings show that ADDF achieves competitive point accuracy and interval prediction performance. Mechanism analyses indicate that the proposed branch-level structure provides clearer interpretability than post-hoc black-box explanations. The framework offers uncertainty-aware forecasting support for sustainable power system operation, including day-ahead scheduling, reserve planning, and energy management. Full article

Objectives: To evaluate the diagnostic performance of automated vertebral trabecular Hounsfield unit (HU) measurements derived from routine fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT) for identifying low bone density. Methods: This retrospective study included 131 consecutive women (mean age, 53.5 ± 9.6 years) undergoing health screening with FDG PET/CT and dual-energy X-ray absorptiometry (DXA) between January 2020 and December 2024. A deep learning-based model (TotalSegmentator) automatically segmented the lumbar vertebrae (L1–L4). HU-based metrics in trabecular regions were calculated, and their correlations with DXA-derived bone mineral density (BMD) were assessed. Diagnostic performance was evaluated using receiver operating characteristic analysis. A multivariable logistic regression model incorporating mean HU, age, and body mass index was developed and internally validated using bootstrap resampling. Results: According to WHO criteria, 47 of 131 participants (35.9%) had low bone density. Mean HU demonstrated strong diagnostic performance (area under the curve [95% confidence interval]: L1, 0.861 [0.800–0.923]; L2, 0.852 [0.788–0.915]; L3, 0.861 [0.800–0.921]; L4, 0.845 [0.781–0.909]). L1 mean HU provided the most balanced performance (sensitivity, 0.851; specificity, 0.750); L3 mean HU was slightly inferior. L1 mean HU was strongly correlated with BMD (r = 0.821, p < 0.001). In multivariable analysis, mean HU independently predicted low bone density (odds ratio: 0.949, p < 0.001). The model achieved an accuracy of 0.786 and demonstrated favorable calibration performance. Conclusions: The automated assessment of vertebral trabecular HU from routine FDG PET/CT provides a reliable and highly efficient method for screening low bone density without additional radiation exposure or cost. Full article

In contemporary society, breast health has become a significant public health concern, particularly among women. According to statistics from the World Health Organization, both the incidence and mortality rates of breast tumors have steadily increased in recent years. Therefore, effective early-stage screening and postoperative monitoring are essential for maintaining breast health. However, conventional clinical diagnostic modalities are typically bulky, operationally complex, and unsuitable for continuous real-time monitoring, which limits their use in portable and everyday health management applications. To address these limitations, this study proposes a machine learning-integrated wearable piezoelectric sensing platform as an auxiliary tool for breast health assessment. The device consists of a PDMS matching layer embedded with flexible silver nanowires, a P(VDF-TrFE) piezoelectric layer, and a multi-channel low-noise signal acquisition circuit. It is capable of acquiring acoustic echo signals from tissue-mimicking environments and automatically evaluating signal validity using a convolutional neural network (CNN). By integrating piezoelectric sensing with deep learning-based signal analysis, the proposed system achieves a signal-to-noise ratio exceeding 70 dB and a real-time classification accuracy above 96% under controlled conditions. These results demonstrate that the platform provides a compact, portable, and intelligent approach for wearable sensing of mechanical heterogeneity and highlight its potential for future development in continuous biomedical monitoring technologies. Full article

To address the significant differences in the structure and mechanical properties of various components of the Shanghai bok choy plug seedling, and the lack of an accurate and reliable discrete element model of the whole plant and key bonding parameters in the simulation of the automatic transplanting process, a 128-cell Shanghai bok choy plug seedling was selected as the research object. Morphological, physical, mechanical, and contact property tests were systematically conducted to obtain the basic parameters of the seedling pot, leaf, petiole, and stem. A whole-plant discrete element model of Shanghai bok choy plug seedling, consisting of the seedling pot, leaf, petiole, and stem, was established using a combined method of component-wise modeling and overall reconstruction. The Hertz–Mindlin (no slip) and Bonding V2 contact models were jointly adopted to characterize interparticle contact, continuous structural behavior, and failure characteristics. Taking the ultimate compressive failure load of the seedling pot, leaf compression density, ultimate bending failure load of the petiole, and ultimate bending failure load of the stem as response indices, significant parameters were screened using the Plackett–Burman test, the optimization ranges were determined through the steepest ascent test, and the key bonding parameters were optimized and calibrated using the Box–Behnken response surface test. The results showed that the relative errors between the simulated and experimental values of the ultimate compressive failure load of the seedling pot, leaf compression density, ultimate bending failure load of the petiole, and ultimate bending failure load of the stem after optimization were 1.19%, 1.13%, 0.99%, and 0.72%, respectively, indicating that the established model can accurately characterize the mechanical response of the constituent parts of Shanghai bok choy plug seedling. The results provide a basis for discrete element simulation of the interaction between Shanghai bok choy plug seedling and key components of automatic transplanting equipment, as well as for the design optimization of automatic transplanting equipment. Full article

Affected by inter-annotator cognitive differences, fatigue effects, and data poisoning, training data inevitably contains a certain proportion of noise, which severely impairs model performance. Traditional manual verification is costly and inefficient, while existing automatic detection methods generally suffer from limited precision, poor interpretability, and insufficient robustness. This paper proposes a noise label detection and correction method based on Bayesian weighted consensus inference. First, an ensemble of multiple lightweight heterogeneous models is constructed, and model prior knowledge and dataset noise are obtained on a clean validation set. Second, the model ensemble predicts noisy samples to extract two-dimensional consensus evidence. Then, prior knowledge and consensus evidence are fused, and the posterior probability of label noise is calculated via Bayesian inference to generate correction suggestions. Finally, high-confidence noisy labels are precisely screened based on the posterior probability threshold. Experimental results on three datasets show that the proposed method achieves a precision of 96.50%, a recall of 98.61%, an F₁-score of 97.54%, and a correction accuracy of 95.53%, with improvements of 5–20% over mainstream methods. With a computational cost comparable to that of basic ensemble methods, the proposed approach achieves a favorable balance among precision, robustness, and interpretability. It thus offers a promising and cost-effective solution for automated quality control of large-scale annotated datasets, especially in text classification tasks. Full article

The essential role of electricity supply for public and private services highlights the need to monitor the stability of power transmission networks during, or immediately after, hazardous events. In the aftermath of calamities, traditional field inspections may be impractical or unsafe, leaving operators without timely information on the condition of critical assets. In this paper, we present and discuss the performance of two automatic Artificial Intelligence (AI)-based models (Multi-Layer Perceptron (MLP) neural network architectures and Support Vector Machine (SVM) model) designed to automatically assess the status of high-voltage transmission towers and power lines through multi-temporal spaceborne Synthetic Aperture Radar (SAR) image analysis. Model development and testing rely on real COSMO-SkyMed Stripmap observations of damaged towers and power lines affected by documented hazardous events across Italy, complemented by simulated tower data generated with a physics-guided, signature-based SAR simulator designed to preserve the observed target-to-background contrast and spatial footprint patterns of real SAR tower signatures. Results indicate that the MLP, trained on either real or simulated data, achieved 100% Overall Accuracy (OA) with no observed false positives or false negatives within the considered visibility-screened real test set, while providing inference times on the order of tenths of milliseconds per target… Computational performance characteristics, operational advantages, and the potential pathway toward satellite on-board porting are discussed to enhance situational awareness and support the prioritisation of interventions during critical events. Full article

With the rapid aging of the global population, constipation has become a major gastrointestinal concern among elderly individuals. Bowel sounds provide a non-invasive acoustic signal for assessing gastrointestinal function, but their automatic analysis remains challenging due to sparsity and non-stationarity. This study proposes a two-stage bowel sound analysis framework based on continuous abdominal recordings. First, a Convolutional Neural Network-MAMBA (CNN-MAMBA) model was used for salient bowel sound detection. Second, a patient-level constipation classification model was developed using multi-view spectral representations and a Convolutional Neural Network-Conformer-Multiple Instance Learning (CNN-Conformer-MIL) architecture. On a held-out test set, the detection model achieved an accuracy of 0.87, an F1-score of 0.78, and a ROC-AUC of 0.93. For patient-level classification under binary Bristol Stool Form Scale (BSFS) grouping, five-fold cross-validation yielded a mean accuracy of 0.665 and an F1-score of 0.755. All BSFS labels were annotated by clinical physicians and temporally aligned with bowel sound recording. Given the modest improvement and cross-validation variability, the patient-level results should be interpreted as preliminary feasibility evidence. These findings suggest that bowel sound analysis may serve as an auxiliary screening or longitudinal monitoring tool rather than a stand-alone diagnostic system. Full article

Emission factor matching is the most critical step in product carbon footprint (PCF) accounting based on life cycle assessment (LCA). However, this step has long been hindered by several major challenges: a lack of standardization, overreliance on expert judgment, inconsistencies in raw data, and complex processing workflows. To address these issues, this study proposes an automated emission factor matching algorithm that combines large language models (LLMs) with semantic retrieval. The algorithm proceeds in two stages: first, an LLM identifies the reference product within the LCA database; then, an embedding model retrieves the most relevant emission factors through high-precision matching. Depending on practical requirements, the algorithm can either automatically select a single best-match factor or rank multiple best-match candidates in descending order of match precision to assist LCA experts in decision-making. We evaluate the algorithm on eight industrial products—tires, cement, ammonium phosphate, wood products, textiles, electronics and electrical appliances, steel, and lithium batteries—using the Ecoinvent 3.10 LCA database. Results demonstrate that the algorithm achieves high precision and low processing latency, significantly outperforming manual expert screening. These findings confirm that the proposed algorithm enables efficient and accurate emission factor matching, thereby providing a reliable technical solution and decision-making pathway for large-scale, automated PCF accounting. Full article

Composed Video Retrieval (CoVR) enables flexible video search by retrieving a target video that reflects a specified modification to a query video. The triplet datasets—consisting of query videos, query text, and target videos—required for model training have been collected manually. Recent studies have explored automatic construction of training triplets for CoVR; however, most existing approaches rely heavily on caption similarity. This limitation is particularly problematic in soccer videos, where identical or highly similar captions can correspond to visually distinct situations, making it difficult to construct triplets with appropriate relationships. To address this issue, this paper proposes a multimodal triplet construction framework specialized for soccer videos. The key idea is to explicitly incorporate visual similarity alongside textual similarity. Specifically, candidate target videos are selected by combining visual similarity with commentary caption filtering, enabling the identification of videos that are visually similar yet semantically different. The semantic difference between videos is then generated as query text using a large language model (LLM) without manual annotation. Furthermore, a multimodal large language model (MLLM) is introduced to estimate whether the generated modification is visually and semantically consistent with the video pair. Rather than replacing human verification, this step provides an automated screening signal to identify potentially unreliable triplets. The experiments show that the proposed framework automatically constructs triplets with reasonable validity under limited human validation. These results demonstrate the potential of scalable triplet construction for CoVR in soccer videos. Full article

Background/Objectives: The dental setting has been suggested as a location for opportunistic diabetes screenings. Diabetes screening is a pathway consisting of several steps that must be completed to reach a diagnosis. Previous research has found that most patients in the dental setting, when offered the opportunity to screen for diabetes, are willing to do so; however, amongst those who are referred for medical follow-up, there is low compliance. If diabetes screening in the dental setting is to be effective, strategies are required to maximise uptake and ensure completion of the screening pathway. Methods: This qualitative study examined participants in a diabetes screening trial held at dental clinics in Victoria, Australia. Semi-structured interviews were conducted by telephone, transcribed and analysed thematically. The themes identified were then deductively mapped onto the Capability, Opportunity, Motivation, Behaviour (COM-B) model and Theoretical Domains Framework (TDF). Results: Ten individuals who were screened for diabetes and referred to their general medical practitioner (GP) for a diabetes diagnosis were interviewed. The themes identified from the interviews were mapped to five COM-B domains: reflective motivation and automatic motivation, social and physical opportunity and psychological capability. These were linked to eight TDF domains associated with issues related to knowledge, environmental context and resources, memory, attention and decision processes, social influences, beliefs about consequences, emotions, and beliefs about capability. Conclusions: This study investigated the determinants influencing individuals’ decision to participate in diabetes screening and comply with referral advice. The results demonstrate the need to increase community knowledge around diabetes and screening for the condition, facilitate risk interpretation, and streamline the referral pathway between oral health professionals (OHP) and GPs. The study provides evidence that can be utilised for the development of future interventions that promote diabetes screening participation and maximise medical follow-up of referred individuals. Full article

Breast cancer screening is a fundamentally probabilistic diagnostic task that requires precise identification of complex imaging characteristics from diverse patient cohorts. Despite improvements in deep learning techniques, current automatic tools are typically trained on well-curated datasets and do not generalize to heterogeneous data, thereby limiting their application. This study aims to address these shortcomings by introducing a more effective and generalizable framework for breast cancer classification that focuses on the stability of features, the learning of complementary representations, and improved decision modeling. The proposed methodology incorporates stability-driven feature extraction (SDFE) with a multi-branch architecture that consists of EfficientNetV2 (Convolutional neural networks (CNNs)), EfficientFormer (Vision transformers (ViTs)), and multi-layer perceptron (MLP)-Mixer models to extract various feature representations. To improve non-linear decision boundaries, it uses a Kolmogorov–Arnold Network (KAN)-based classification head and selects the most credible prediction via an adaptive voting mechanism. This model is trained using patient-level splitting on the VinDr-Mammo dataset, evaluated using five-fold cross-validation, and subsequently externally validated on the CBIS-DDSM dataset. Experimental findings demonstrate the consistent performance of the proposed model, with accuracies of 94.5% in cross-validation, 93.3% on the VinDr-Mammo test set, and 94.6% on CBIS-DDSM, surpassing other recent state-of-the-art solutions. It demonstrates enhanced robustness and cross-dataset generalization, offering a scalable, consistent framework for breast cancer classification that supports the development of computer-aided diagnostic systems. Full article

Down quality is the core evaluation indicator of thermal insulation products, and its grade determination strictly complies with the down content index specified in the national standard GB/T 17685-2016 Feather and Down. Traditional down quality inspection adopts manual sorting and weighing methods, which are plagued by low efficiency, strong subjectivity and high error rates, thereby restricting the intelligent upgrading of the down industry. This study aims to develop an automatic down detection and quantitative grading method conforming to national standards based on deep learning. A down dataset consisting of 632 RGB images is constructed, with each image containing 5–10 individual down samples and covering five categories: mature down clusters, immature down clusters, down filaments, feathers, and yellow-tail down. Three mainstream frameworks including YOLOv8, YOLOv11 and YOLOv26 are trained for performance comparison. Precision, recall, mAP@50 and mAP@50-95 are adopted as evaluation metrics. In addition, this paper proposes a research idea for down content calculation and automatic classification and grading of down quality in accordance with relevant national standards. The experimental results demonstrate that the latest models do not necessarily achieve the optimal performance. The newly released YOLOv26n and YOLOv26m exhibit relatively low accuracy in the down detection task, with mAP@50 values of only 0.98556 and 0.99077, and recall rates of 0.95032 and 0.97848, respectively, failing to outperform their previous-generation counterparts. In contrast, YOLOv11n achieves the best comprehensive performance, with an mAP@50 of 0.99416, a precision of 0.99544, a recall of 0.99722, and an mAP@50-95 of 0.63464. Meanwhile, the model has only 2.58 M parameters, a computational complexity of 6.3 GFLOPs, and a single training time of approximately 6.7 min, achieving an optimal balance between detection accuracy and computational efficiency. All models show the highest detection accuracy for mature down clusters and yellow-tailed down, while slight confusion exists between immature down clusters and down filaments. This study verifies the feasibility of the YOLO series models in down quality inspection in accordance with national standards, and reveals that model architecture iteration does not necessarily lead to performance improvement on specific industrial datasets. The lightweight and robustly designed YOLOv11n presents greater practical value. The intelligent detection scheme proposed in this paper can assist in optimizing the traditional manual quality inspection workflow, alleviating the burden of manual counting and reducing subjective errors. It provides new ideas and technical references for the rapid screening and objective determination of down quality. Furthermore, the proposed research framework for automatic classification and grading of down quality is expected to promote the development of down quality inspection toward standardization, intelligence, and automation in the future. Full article

In recent years, augmented reality has become a popular method of spatial data visualization, both via the most popular and basic plane-based method and more advanced automatic positioning of visualizations based on predefined real-world locations. The aim of this study is to provide new insights into geolocated 3D visualizations in AR using a visual positioning system (VPS). VPS technology enables the creation of visualizations that can be displayed with high accuracy directly on a specific area of interest. This approach is especially well-suited to cultural heritage preservation, as it can be used to visualize destroyed buildings or archaeological sites. The result of the study is a mobile application created using the Unity game engine, which allows users to access AR visualizations as well as additional context in the form of pop-up texts or photographs. Thanks to the display of AR visualization directly at the chosen location, the user can better understand the context of the whole scene. This is because it is a more immersive experience than simply viewing a 3D model on a computer or mobile phone screen. Full article

Background: Viral Pneumonia and Tuberculosis continue to represent a significant burden on global public health, relying heavily on chest X-rays for screening and diagnosis. Although deep learning systems offer promising diagnostic support, the traditional manual tuning of hyperparameters for Convolutional Neural Networks is often inefficient and computationally expensive, frequently resulting in suboptimal or overly heavy architectures. Methods: To address these challenges, this study proposes a hybrid framework that employs metaheuristic algorithms, specifically the Whale Optimization Algorithm, Grey Wolf Optimizer, and Cuckoo Search to automatically optimize the architecture and training parameters of a custom neural network for the multi-class classification of Normal, Viral Pneumonia, and Tuberculosis cases. The proposed approach was evaluated using a rigorous stratified k-fold cross-validation protocol on a balanced, multi-source dataset. Results: The experimental results demonstrate that the model optimized by the Whale Optimization Algorithm statistically outperforms manually configured baselines, achieving the highest diagnostic accuracy and specificity. Furthermore, a critical finding of this research is the substantial improvement in computational efficiency; the automated optimization reduced the computational load by approximately 74% and the storage requirements by 63%, making the model viable for deployment in resource-constrained environments. Conclusions: Finally, to ensure clinical reliability, the decision-making process was validated using Gradient-weighted Class Activation Mapping, which confirmed that the network successfully learns to identify clinically relevant pulmonary structures while ignoring confounding artifacts. Full article

Electric vehicles introduce new considerations for pedestrian safety because their lower operating noise at low speeds may reduce pedestrian detectability in urban traffic environments. This study proposes a simulation-based integrated active–passive pedestrian protection framework for electric vehicles by linking automatic emergency braking, active hood deployment, and post-crash head injury assessment. A total of 688 valid pedestrian–vehicle crash records from the National Highway Traffic Safety Administration database were analyzed, and 5 representative pedestrian crash scenarios were constructed through clustering-informed scenario screening and a benchmark pedestrian AEB scenario. The scenarios were reconstructed in a PreScan–Simulink co-simulation environment to evaluate a time-to-collision-based AEB strategy, while the active hood system was assessed using multi-body dynamics simulation and finite element head impact analysis. The AEB results showed that three scenarios were avoided before pedestrian contact, whereas two remained unavoidable, with residual impact speeds of approximately 31.5 km/h and 46 km/h. The hood reached a stable deployed posture within approximately 0.1 s under the modeled conditions. The HIC₁₅ results at eight selected impact points showed that speed reduction and hood deployment generally reduced head injury metrics, but full compliance with the reference HIC₁₅ threshold of 1000 was not achieved at all points. These findings suggest that the proposed strategy can improve simulated pedestrian head protection performance under selected electric vehicle crash scenarios, while further structural optimization, experimental validation, and cost–benefit assessments are still required. Full article