Search Results (940)

Background: Chest X-ray (CXR) is widely used for the assessment of thoracic diseases, yet automated multi-label interpretation remains challenging due to subtle visual patterns, overlapping anatomical structures, and frequent co-occurrence of abnormalities. While recent deep learning models have shown strong performance, limitations in interpretability, anatomical awareness, and robustness continue to hinder their clinical adoption. Methods: The proposed framework employs a hybrid ConvNeXtV2–Vision Transformer (ViT) architecture that combines convolutional feature extraction for capturing fine-grained local patterns with transformer-based global reasoning to model long-range contextual dependencies. The model is trained exclusively using image-level annotations. In addition to classification, three complementary post hoc components are integrated to enhance model trust and interpretability. A segmentation-aware Gradient-weighted class activation mapping (Grad-CAM) module leverages CheXmask lung and heart segmentations to highlight anatomically relevant regions and quantify predictive evidence inside and outside the lungs. An ontology-driven neuro-symbolic reasoning layer translates Grad-CAM activations into structured, rule-based explanations aligned with clinical concepts such as “basal effusion” and “enlarged cardiac silhouette”. Furthermore, a lightweight out-of-distribution (OOD) detection module based on confidence scores, energy scores, and Mahalanobis distance scores is employed to identify inputs that deviate from the training distribution. Results: On the VinBigData test set, the model achieved a macro-AUROC of 0.9525 and a Micro AUROC of 0.9777 when trained solely with image-level annotations. External evaluation further demonstrated strong generalisation, yielding macro-AUROC scores of 0.9106 on NIH ChestXray14 and 0.8487 on CheXpert (frontal views). Both Grad-CAM visualisations and ontology-based reasoning remained coherent on unseen data, while the OOD module successfully flagged non-thoracic images. Conclusions: Overall, the proposed approach demonstrates that hybrid convolutional neural network (CNN)–vision transformer (ViT) architectures, combined with anatomy-aware explainability and symbolic reasoning, can support automated chest X-ray diagnosis in a manner that is accurate, transparent, and safety-aware. Full article

Background/Objectives: Accurate identification of dental malocclusions from routine clinical photographs can be time-consuming and subject to interobserver variability. A YOLOv11-based deep learning approach is presented and evaluated for automatic malocclusion detection on routine intraoral photographs, testing the hypothesis that training on a structured annotation protocol enables reliable detection of multiple clinically relevant malocclusions. Methods: An anonymized dataset of 5854 intraoral photographs (frontal occlusion; right/left buccal; maxillary/mandibular occlusal) was labeled according to standardized instructions derived from the Index of Orthodontic Treatment Need (IOTN) A total of 17 clinically relevant classes were annotated with bounding boxes. Due to an insufficient number of examples, two malocclusions (transposition and non-occlusion) were excluded from our quantitative analysis. A YOLOv11 model was trained with augmented data and evaluated on a held-out test set using mean average precision at IoU 0.5 (mAP50), macro precision (macro-P), and macro recall (macro-R). Results: Across 15 analyzed classes, the model achieved 87.8% mAP50, 76.9% macro-P, and 86.1% macro-R. The highest per-class AP₅₀ was observed for Deep bite (98.8%), Diastema (97.9%), Angle Class II canine (97.5%), Anterior open bite (92.8%), Midline shift (91.8%), Angle Class II molar (91.1%), Spacing (91%), and Crowding (90.1%). Moderate performance included Anterior crossbite (88.3%), Angle Class III molar (87.4%), Head bite (82.7%), and Posterior open bite (80.2%). Lower values were seen for Angle Class III canine (76%), Posterior crossbite (75.6%), and Big overjet (75.3%). Precision–recall trends indicate earlier precision drop-off for posterior/transverse classes and comparatively more missed detections in Posterior crossbite, whereas Big overjet exhibited more false positives at the chosen threshold. Conclusion: A YOLOv11-based deep learning system can accurately detect several clinically salient malocclusions on routine intraoral photographs, supporting efficient screening and standardized documentation. Performance gaps align with limited examples and visualization constraints in posterior regions. Larger, multi-center datasets, protocol standardization, quantitative metrics, and multimodal inputs may further improve robustness. Full article

This paper examines the contrasting yet complementary approaches of the Constrained Disorder Principle (CDP) and Stefan Hell’s deterministic optical nanoscopy for managing noise in complex systems. The CDP suggests that controlled disorder within dynamic boundaries is crucial for optimal system function, particularly in biological contexts, where variability acts as an adaptive mechanism rather than being merely a measurement error. In contrast, Hell’s recent breakthrough in nanoscopy demonstrates that engineered diffraction minima can achieve sub-nanometer resolution without relying on stochastic (random) molecular switching, thereby replacing randomness with deterministic measurement precision. Philosophically, these two approaches are distinct: the CDP views noise as functionally necessary, while Hell’s method seeks to overcome noise limitations. However, both frameworks address complementary aspects of information extraction. The primary goal of microscopy is to provide information about structures, thereby facilitating a better understanding of their functionality. Noise is inherent to biological structures and functions and is part of the information in complex systems. This manuscript achieves integration through three specific contributions: (1) a mathematical framework combining CDP variability bounds with Hell’s precision measurements, validated through Monte Carlo simulations showing 15–30% precision improvements; (2) computational demonstrations with N = 10,000 trials quantifying performance under varying biological noise regimes; and (3) practical protocols for experimental implementation, including calibration procedures and real-time parameter optimization. The CDP provides a theoretical understanding of variability patterns at the system level, while Hell’s technique offers precision tools at the molecular level for validation. Integrating these approaches enables multi-scale analysis, allowing for deterministic measurements to accurately quantify the functional variability that the CDP theory predicts is vital for system health. This synthesis opens up new possibilities for adaptive imaging systems that maintain biologically meaningful noise while achieving unprecedented measurement precision. Specific applications include cancer diagnostics through chromosomal organization variability, neurodegenerative disease monitoring via protein aggregation disorder patterns, and drug screening by assessing cellular response heterogeneity. The framework comprises machine learning integration pathways for automated recognition of variability patterns and adaptive acquisition strategies. Full article

Non-obstructive azoospermia (NOA) is a severe male infertility condition characterized by extremely low or absent sperm production. In microdissection testicular sperm extraction (Micro-TESE) procedures for NOA, embryologists must manually search through testicular tissue under a microscope for rare sperm, a process that can take 1.8–7.5 h and impose significant fatigue and burden. This paper presents an augmented reality (AR) microscope system with AI-based image analysis to accelerate sperm retrieval in Micro-TESE. The proposed system integrates a deep learning model (YOLOv5) for real-time sperm detection in microscope images, a multi-object tracker (DeepSORT) for continuous sperm tracking, and a velocity calculation module for sperm motility analysis. Detected sperm positions and motility metrics are overlaid in the microscope’s eyepiece view via a microdisplay, providing immediate visual guidance to the embryologist. In experiments on seminiferous tubule sample images, the YOLOv5 model achieved a precision of 0.81 and recall of 0.52, outperforming previous classical methods in accuracy and speed. The AR interface allowed an operator to find sperm faster, roughly doubling the sperm detection rate (66.9% vs. 30.8%). These results demonstrate that the AR microscope system can significantly aid embryologists by highlighting sperm in real time and potentially shorten Micro-TESE procedure times. This application of AR and AI in sperm retrieval shows promise for improving outcomes in assisted reproductive technology. Full article

Exposure models are essential for seismic risk assessment to determine environmental vulnerabilities during earthquakes. However, developing these models at scale is challenging because it relies on manual inspection of buildings, which increases costs and introduces significant delays. Developing fast, consistent, and easy-to-deploy automated methods to support this process has become a priority. In this study, we investigate the use of deep learning to accelerate the classification of architectural and structural attributes from street-view imagery. Using the Alvalade dataset, which contains 4007 buildings annotated with 10 multi-class attributes, we evaluated the performance of multiple architecture types. Our analysis shows that deep learning models can successfully extract key structural features, achieving an average macro accuracy of 57%, and a Precision, Recall, and F1-score of 61%, 57%, and 56%, respectively. We also show that prediction quality is further improved by leveraging multi-view imagery of the target buildings. These results demonstrate that deep learning can be an effective solution to reduce the manual effort required for the development of reliable large-scale exposure models, offering a practical solution toward more efficient seismic risk assessment. Full article

Accurate assessment of rice resistance to Sogatella furcifera (Horváth) is essential for breeding insect-resistant cultivars. Traditional assessment methods rely on manual scoring of damage severity, which is subjective and inefficient. To overcome these limitations, this study proposes an automated resistance evaluation approach based on multi-view 3D reconstruction and deep learning–based point cloud segmentation. Multi-view videos of rice materials with different resistance levels were collected over time and processed using Structure from Motion (SfM) and Multi-View Stereo (MVS) to reconstruct high-quality 3D point clouds. A well-annotated “3D Rice WBPH Damage” dataset comprising 174 samples (15 rice materials, three replicates each, 45 pots) was established, where each sample corresponds to a reconstructed 3D point cloud from a video sequence. A comparative study of various point cloud semantic segmentation models, including PointNet, PointNet++, ShellNet, and PointCNN, revealed that the PointNet++ (MSG) model, which employs a Multi-Scale Grouping strategy, demonstrated the best performance in segmenting complex damage symptoms. To further accurately quantify the severity of damage, an adaptive point cloud dimensionality reduction method was proposed, which effectively mitigates the interference of leaf shrinkage on damage assessment. Experimental results demonstrated a strong correlation (R² = 0.95) between automated and manual evaluations, achieving accuracies of 86.67% and 93.33% at the sample and material levels, respectively. This work provides an objective, efficient, and scalable solution for evaluating rice resistance to S. furcifera, offering promising applications in crop resistance breeding. Full article

In intelligent construction and BIM–Reality integration applications, high-quality, large-scale construction scene point cloud data with component-level semantic annotations constitute a fundamental basis for three-dimensional semantic understanding and automated analysis. However, point clouds acquired from real construction sites commonly suffer from high labeling costs, severe occlusion, and unstable data distributions. Existing public datasets remain insufficient in terms of scale, component coverage, and annotation consistency, limiting their suitability for data-driven approaches. To address these challenges, this paper constructs and releases a BIM-derived synthetic construction scene point cloud dataset, termed the Synthetic Point Cloud (SPC), targeting component-level point cloud semantic segmentation and related research tasks.The dataset is generated from publicly available BIM models through physics-based virtual LiDAR scanning, producing multi-view and multi-density three-dimensional point clouds while automatically inheriting component-level semantic labels from BIM without any manual intervention. The SPC dataset comprises 132 virtual scanning scenes, with an overall scale of approximately $8.75\times {10}^9$ points, covering typical construction components such as walls, columns, beams, and slabs. By systematically configuring scanning viewpoints, sampling densities, and occlusion conditions, the dataset introduces rich geometric and spatial distribution diversity. This paper presents a comprehensive description of the SPC data generation pipeline, semantic mapping strategy, virtual scanning configurations, and data organization scheme, followed by statistical analysis and technical validation in terms of point cloud scale evolution, spatial coverage characteristics, and component-wise semantic distributions. Furthermore, baseline experiments on component-level point cloud semantic segmentation are provided. The results demonstrate that models trained solely on the SPC dataset can achieve stable and engineering-meaningful component-level predictions on real construction point clouds, validating the dataset’s usability in virtual-to-real research scenarios. As a scalable and reproducible BIM-derived point cloud resource, the SPC dataset offers a unified data foundation and experimental support for research on construction scene point cloud semantic segmentation, virtual-to-real transfer learning, scan-to-BIM updating, and intelligent construction monitoring. Full article

While text-to-image and customized generation methods demonstrate strong capabilities in single-image generation, they fall short in supporting immersive applications that require coherent 360° panoramas. Conversely, existing panorama generation models lack customization capabilities. In panoramic scenes, reference objects often appear as minor background elements and may be multiple in number, while reference images across different views exhibit weak correlations. To address these challenges, we propose a diffusion-based framework for customized multi-view image generation. Our approach introduces a decoupled feature injection mechanism within a dual-UNet architecture to handle weakly correlated reference images, effectively integrating spatial information by concurrently feeding both reference images and noise into the denoising branch. A hybrid attention mechanism enables deep fusion of reference features and multi-view representations. Furthermore, a data augmentation strategy facilitates viewpoint-adaptive pose adjustments, and panoramic coordinates are employed to guide multi-view attention. The experimental results demonstrate our model’s effectiveness in generating coherent, high-quality customized multi-view images. Full article

To address the challenge of bearing load monitoring in elastically supported marine shafting systems, a multi-data fusion-based prediction model is constructed. In view of the small-sample nature of measured bearing load data, transfer learning is adopted to migrate the physical relationships embedded in finite element simulations to the measurement domain. A limited number of actual samples are used to correct the simulation data, forming a high-fidelity hybrid training set. The system—supported by air-spring isolators mounted on the raft—is divided into multiple sub-regions according to their spatial layout, establishing local mappings from air-spring pressure variations to bearing load increments to reduce model complexity. On this basis, a Stacking ensemble learning framework is further incorporated into the prediction model to integrate multi-source information such as air-spring pressure and raft strain, thereby enriching the model’s information acquisition and improving prediction accuracy. Experimental results show that the proposed transfer learning-based multi-sub-region bearing load prediction model performs significantly better than the full-parameter input model. Furthermore, the strain-enhanced Stacking-based multi-data fusion bearing load prediction model improves the characterization of shafting features and reduces the maximum prediction error. The proposed multi-data fusion modeling strategy offers a viable approach for condition monitoring and intelligent maintenance of marine shafting systems. Full article

Power-equipment fault detection is challenging in real-world inspections due to subtle defect cues and cluttered backgrounds. This paper proposes an improved YOLOv12-based framework for multi-class power defect detection. We introduce a Prior-Guided Region Attention (PG-RA) module and design a Lightweight Residual Efficient Layer Aggregation Network (LR-RELAN). In addition, we develop a Dual-Spectrum Adaptive Fusion Loss (DSAF Loss) function to jointly improve classification confidence and bounding box regression consistency, enabling more robust learning under complex scenes. To support defect-oriented visual analytics and system interpretability, the framework further constructs Region of Interest (ROI) pseudo point clouds from detection outputs and compares two denoising strategies, Statistical Outlier Removal (SOR) and Radius Outlier Removal (ROR). A Python-based graphical prototype integrates image import, defect detection, ROI pseudo point cloud construction, denoising, 3D visualization, and result archiving into a unified workflow. Experimental results demonstrate that the proposed method improves detection accuracy and robustness while maintaining real-time performance, and the ROI pseudo point cloud module provides an intuitive auxiliary view for defect-structure inspection in practical applications. Full article

Three-dimensional visual grounding is a core technology that identifies specific objects within complex 3D scenes based on natural language instructions, enhancing human–machine interactions in robotics and augmented reality domains. Traditional approaches have focused on supervised learning, which relies on annotated data; however, zero-shot methodologies are emerging due to the high costs of data construction and limitations in generalization. SeeGround achieves state-of-the-art performance by integrating 2D rendered images and spatial text descriptions. Nevertheless, SeeGround exhibits vulnerabilities in clearly discerning relative depth relationships owing to its implicit depth representations in 2D views. This study proposes the relational depth text (RDT) technique to overcome these limitations, utilizing a Monocular Depth Estimation model to extract depth maps from rendered 2D images and applying the K-Nearest Neighbors algorithm to convert inter-object relative depth relations into natural language descriptions, thereby incorporating them into Vision–Language Model (VLM) prompts. This method distinguishes itself by augmenting spatial reasoning capabilities while preserving SeeGround’s existing pipeline, demonstrating a 3.54% improvement in the Acc@0.25 metric on the Nr3D dataset in a 7B VLM environment that is approximately 10.3 times lighter than the original model, along with a 6.74% increase in Unique cases on the ScanRefer dataset, albeit with a 1.70% decline in Multiple cases. The proposed technique enhances the robustness of grounding through viewpoint anchoring and candidate discrimination in complex query scenarios, and is expected to improve efficiency in practical applications through future multi-view fusion and conditional execution optimizations. Full article

Collaborative sensing between low-altitude remote sensing and ground-based mobile mapping lays the theoretical foundation for multi-platform 3D data fusion. However, point clouds collected from Airborne Laser Scanners (ALSs) remain scarce due to high acquisition and annotation costs. In contrast, while autonomous driving datasets are more accessible, dense annotation remains a significant bottleneck. To address this, we propose Ground to Altitude (GTA), a weakly supervised domain generalization (DG) framework. GTA leverages sparse autonomous driving data to learn robust representations, enabling reliable segmentation on airborne point clouds under zero-label conditions. Specifically, we tackle cross-platform discrepancies through progressive domain-aware augmentation (PDA) and cross-scale semantic alignment (CSA). For PDA, we design a distance-guided dynamic upsampling strategy to approximate airborne point density and a cross-view augmentation scheme to model viewpoint variations. For CSA, we impose cross-domain feature consistency and contrastive regularization to enhance robustness against perturbations. A progressive training pipeline is further employed to maximize the utility of limited annotations and abundant unlabeled data. Our study reveals the limitations of existing DG methods in cross-platform scenarios. Extensive experiments demonstrate that GTA achieves state-of-the-art (SOTA) performance. Notably, under the challenging 0.1% supervision setting, our method achieves a 6.36% improvement in mIoU over the baseline on the SemanticKITTI → DALES benchmark, demonstrating significant gains across diverse categories beyond just structural objects. Full article

By enhancing transient fidelity before geometric inversion, this work revisits the classical LCT-based non line-of-sight (NLOS)imaging paradigm and establishes a unified Bayesian sparse-enhancement framework for reconstructing hidden objects under photon-starved and hardware-limited conditions. We introduce sparse Bayesian learning (SBL) as a dedicated front-end transient restoration module, leveraging adaptive sparsity modeling to suppress background fluctuations while preserving physically consistent multipath returns. This lightweight and geometry-agnostic design enables seamless integration into existing LCT processing pipelines, granting the framework strong compatibility with diverse acquisition configurations. Comprehensive simulations and experiments on complex reflective targets demonstrate significant improvements in spatial resolution, boundary sharpness, and robustness to IRF-induced temporal blurring compared with traditional LCT and f-k migration methods. The results validate that transient quality remains a critical bottleneck in practical NLOS deployment, and addressing it via probabilistic sparsity inference offers a scalable and computationally affordable pathway toward stable, high-fidelity NLOS reconstruction. This study provides an effective signal-domain enhancement solution that strengthens the practicality of NLOS imaging in real-world environments, paving the way for future extensions toward dynamic scenes, multi-view fusion, and high-throughput computational sensing. Full article

Using Machine Learning (ML) in educational management transforms higher education strategy. This study examines students’ views on machine learning (ML) technologies and how they might be used to plan, monitor, and predict student performance. The Faculty of Industrial Engineering and Robotics surveyed 118 third-year undergraduates. It featured closed- and open-ended questions to collect quantitative and qualitative data. Descriptive statistics showed broad patterns, inferential tests (Chi-square, t-test, ANOVA) showed group differences, regression models predicted school outcomes, and exploratory factor analysis (EFA) and clustering found hidden attitudes and student profiles. A multi-method quantitative approach combining descriptive statistics, inferential tests, regression modeling, and exploratory techniques (EFA and clustering) was employed. The findings show that most students realize that ML may help them be more productive, adapt their study pathways, and learn about the future. Concerns remain regarding its accuracy, overreliance, and morality. The findings indicate that ML can both support and challenge educational management, depending on how responsibly it is implemented. Results show that institutions may utilize ML as a strategic tool to boost academic progress and make better judgments, provided they incorporate it responsibly and follow ethical rules and training. Full article

Wearable multi-modal body fluid monitoring enables continuous, non-invasive, and context-aware assessment of human physiology. By integrating biochemical and physical information across multiple modalities, wearable systems overcome the limitations of single-marker sensing and provide a more holistic view of dynamic health states. This review offers a system-level overview of recent advances in multi-modal body fluid monitoring, structured into three hierarchical dimensions. We first examine sensing-combination strategies such as multi-marker analysis within single fluids, coupling biochemical signals with bioelectrical, mechanical, or thermal parameters, and emerging multi-fluid acquisition to improve analytical accuracy and physiological relevance. Next, we discuss platform-integration mechanisms based on biochemical, physical, and hybrid sensing principles, along with monolithic and modular architectures enabled by flexible electronics, microfluidics, microneedles, and smart textiles. Finally, the data-processing patterns are analyzed, involving cross-modal calibration, machine learning inference, and multi-level data fusion to enhance data reliability and support personalized and predictive healthcare. Beyond summarizing technical advances, this review establishes a comprehensive framework that moves beyond isolated signal acquisition or simple metric aggregation toward holistic physiological interpretation. It guides the development of next-generation wearable multi-modal body fluid monitoring systems that overcome the challenges of high integration, miniaturization, and personalized medical applications. Full article