Search Results (174)

Accurate classification of brain tumors in medical imaging is crucial to ensure reliable diagnoses and effective treatment planning. This study introduces a novel double ensemble framework that synergistically combines pre-trained Deep Learning (DL) models for feature extraction with optimized Machine Learning (ML) classifiers for robust classification. The framework incorporates comprehensive preprocessing and data augmentation of brain Magnetic Resonance Images (MRIs), followed by deep feature extraction based on transfer learning using pre-trained Vision Transformer (ViT) networks. The novelty of our approach lies in its dual-level ensemble strategy: we employ a feature-level ensemble, which integrates deep features from the top-performing ViT models, and a classifier-level ensemble, which aggregates predictions from various hyperparameter-optimized ML classifiers. Experiments on two public MRI brain tumor datasets from Kaggle demonstrate that this approach significantly surpasses state-of-the-art methods, underscoring the importance of feature and classifier fusion. The proposed methodology also highlights the critical roles that hyperparameter optimization and advanced preprocessing techniques can play in improving the diagnostic accuracy and reliability of medical image analysis, advancing the integration of DL and ML in this vital, clinically relevant task. Full article

Background and Clinical Significance: Carcinoid tumors are rare, slow-growing neuroendocrine cell neoplasms that typically affect the gastrointestinal tract. While metastasis may occur, it most commonly occurs in the liver, and orbital metastasis is extremely rare, especially while on systemic somatostatin suppression. Case Presentation: A 57-year-old man with a history of gastrointestinal carcinoid tumor treated with lanreotide for 5 years presented with a left proptotic, red eye and double vision for several months. Clinical examination revealed left proptosis, supraduction deficit, lower lid retraction, and dilated episcleral vessels inferiorly. Magnetic resonance imaging demonstrated a 1.8 cm enhancing lesion centered within the left inferior rectus muscle. Left orbitotomy and biopsy were performed, which confirmed metastatic carcinoid tumor. He will undergo localized orbital radiation and substitution of lanreotide with systemic chemotherapy. Conclusions: Orbital metastasis of carcinoid tumor is extremely uncommon. Given its rarity, diagnosis may be challenging. In patients presenting with ocular complaints including chronic red eye, double vision, proptosis, and mass effect with a prior history of neuroendocrine cancer, a high index of suspicion for orbital metastasis is necessary with timely workup and treatment even if the disease has been otherwise well-controlled with somatostatin analogs. Full article

Background/Objectives: Hypertension is a growing public health concern in Saudi Arabia, driven by rapid socioeconomic changes. This study investigated the interplay between habitual, behavioral, and dietary risk factors associated with hypertension among Saudi adults. Methods: A cross-sectional survey was conducted among 3312 Saudi adults using multistage stratified random sampling. The data were collected via validated questionnaires assessing sociodemographic, anthropometric indicators, lifestyle behaviors, dietary patterns, and medical history. Hypertension status was determined through self-reported diagnosis. Bivariate analyses and multiple logistic regression identified independent predictors (p < 0.05). Results: Hypertension prevalence was 13% (mean age: 34 ± 15 years; 50% male). The strongest predictors were age (OR = 1.08/year; 95% CI: 1.07–1.10; p < 0.001), increased body mass index (OR = 1.03; 95% CI: 1.01–1.06; p = 0.011), smoking (OR = 1.55; 95% CI: 1.04–2.29; p = 0.030), and family history of hypertension (OR = 7.71; 95% CI: 5.61–10.75; p < 0.001). Participants with diabetes mellitus had 89% higher odds of hypertension (OR = 1.89; 95% CI: 1.42–2.51; p < 0.001), and those with dyslipidemia had more than double the odds (OR = 2.45; 95% CI: 1.38–4.22; p = 0.002). Protective factors included higher income (≥15,000 SAR; OR = 0.54; 95% CI: 0.36–0.81; p = 0.003) and regular whole grain consumption (OR = 0.60; 95% CI: 0.46–0.77; p < 0.001). Conclusions: Hypertension risk in Saudi adults is shaped by age, obesity, smoking, comorbid metabolic conditions (diabetes and dyslipidemia), and genetic pre-disposition. In contrast, higher income and whole grain intake may offer protection. These findings underscore the need for comprehensive prevention strategies that address both lifestyle and cardiometabolic comorbidities, in alignment with Saudi Vision 2030 health priorities. Full article

The aim of this study was to address the issue of significant performance degradation in existing defogging algorithms under extreme fog conditions. Traditional Taylor series-based deformable convolutions are limited by local approximation errors, while the heavy-tailed characteristics of the Cauchy distribution can more successfully model outliers in fog images. The following improvements are made: (1) A displacement generator based on the inverse cumulative distribution function (ICDF) of the Cauchy distribution is designed to transform uniform noise into sampling points with a long-tailed distribution. A novel double-peak Cauchy ICDF is proposed to dynamically balance the heavy-tailed characteristics of the Cauchy ICDF, enhancing the modeling capability for sudden changes in fog concentration. (2) An innovative Cauchy–Gaussian fusion module is proposed to dynamically learn and generate hybrid coefficients, combining the complementary advantages of the two distributions to dynamically balance the representation of smooth regions and edge details. (3) Tree-based multi-path and cross-resolution feature aggregation is introduced, achieving local–global feature adaptive fusion through adjustable window sizes (3/5/7/11) for parallel paths. Experiments on the RESIDE dataset demonstrate that the proposed method achieves a 2.26 dB improvement in the peak signal-to-noise ratio compared to that obtained with the TaylorV2 expansion attention mechanism, with an improvement of 0.88 dB in heavily hazy regions (fog concentration > 0.8). Ablation studies validate the effectiveness of Cauchy distribution convolution in handling dense fog and conventional lighting conditions. This study provides a new theoretical perspective for modeling in computer vision tasks, introducing a novel attention mechanism and multi-path encoding approach. Full article

Vision-based navigation is a common solution for the critical challenge of GPS-denied Unmanned Aerial Vehicle (UAV) operation, but a research gap remains in the autonomous discovery of robust landmarks from aerial survey imagery needed for such systems. In this work, we propose a framework to fill this gap by identifying visually distinctive urban buildings from aerial survey imagery and curating them into a landmark database for GPS-free UAV localization. The proposed framework constructs semantically rich embeddings using intermediate layers from a pre-trained YOLOv11n-seg segmentation network. This novel technique requires no additional training. An unsupervised landmark selection strategy, based on the Isolation Forest algorithm, then identifies objects with statistically unique embeddings. Experimental validation on the VPAIR aerial-to-aerial benchmark shows that the proposed max-pooled embeddings, assembled from selected layers, significantly improve retrieval performance. The top-1 retrieval accuracy for landmarks more than doubled compared to typical buildings (0.53 vs. 0.31), and a Recall@5 of 0.70 is achieved for landmarks. Overall, this study demonstrates that unsupervised outlier selection in a carefully constructed embedding space yields a highly discriminative, computation-friendly set of landmarks suitable for real-time, robust UAV navigation. Full article

Visual field (VF) testing remains a cornerstone in assessing retinal function by measuring how well different parts of the retina detect light. It is essential for early detection, monitoring, and management of many retinal diseases. By mapping retinal sensitivity, VF exams can reveal functional loss before structural changes become visible. This review summarizes how VF testing is applied across key conditions: hydroxychloroquine (HCQ) retinopathy, age-related macular degeneration (AMD), diabetic retinopathy (DR) and macular edema (DME), and inherited disorders including inherited dystrophies such as retinitis pigmentosa (RP). Traditional methods like the Goldmann kinetic perimetry and simple tools such as the Amsler grid help identify large or central VF defects. Automated perimetry (e.g., Humphrey Field Analyzer) provides detailed, quantitative data critical for detecting subtle paracentral scotomas in HCQ retinopathy and central vision loss in AMD. Frequency-doubling technology (FDT) reveals early neural deficits in DR before blood vessel changes appear. Microperimetry offers precise, localized sensitivity maps for macular diseases. Despite its value, VF testing faces challenges including patient fatigue, variability in responses, and interpretation of unreliable results. Recent advances in artificial intelligence, virtual reality perimetry, and home-based perimetry systems are improving test accuracy, accessibility, and patient engagement. Integrating VF exams with these emerging technologies promises more personalized care, earlier intervention, and better long-term outcomes for patients with retinal disease. Full article

Rain streaks degrade the quality of optical images and seriously affect the effectiveness of subsequent vision-based algorithms. Although the applications of a convolutional neural network (CNN) and self-attention mechanism (SA) in single image deraining have shown great success, there are still unresolved issues regarding the deraining performance and the large computational load. The work in this paper fully coordinates and utilizes the advantages between CNN and SA and proposes a hybrid content-aware deraining network (CAD) to reduce complexity and generate high-quality results. Specifically, we construct the CADBlock, including the content-aware convolution and attention mixer module (CAMM) and the multi-scale double-gated feed-forward module (MDFM). In CAMM, the attention mechanism is used for intricate windows to generate abundant features and simple convolution is used for plain windows to reduce computational costs. In MDFM, multi-scale spatial features are double-gated fused to preserve local detail features and enhance image restoration capabilities. Furthermore, a four-token contextual attention module (FTCA) is introduced to explore the content information among neighbor keys to improve the representation ability. Both qualitative and quantitative validations on synthetic and real-world rain images demonstrate that the proposed CAD can achieve a competitive deraining performance. Full article

China’s accelerating industrialization has led to worsening air pollution, characterized by recurrent haze episodes. The accurate quantification of PM_2.5 distribution is crucial for air quality assessment and public health management. Although traditional prediction models can effectively identify PM_2.5 concentration fluctuations with moderate accuracy, their dependence relies heavily on extensive ground-based monitoring station data, limiting their applicability in areas with sparse monitoring coverage. To address this limitation, this study proposes a novel algorithm for high-precision PM_2.5 concentration prediction, termed VDMS (Vision Transformer with DLSTM Multi-Head Self-Attention and Self-supervision). Based on the traditional Vision Transformer (ViT) architecture, VDMS incorporates a Double-Layered Long Short-Term Memory (DLSTM) network and a Multi-Head Self-Attention mechanism to enhance the model’s capacity to capture temporal sequence features and global dependencies. These enhancements contribute to greater stability and robustness in feature representation, ultimately improving prediction performance. Cross-validation experimental results show that the VDMS model outperforms benchmark models in PM_2.5 concentration prediction tasks, achieving a coefficient of determination (R²) of 0.93, a root mean square error (RMSE) of 4.05 μg/m³, and a mean absolute error (MAE) of 3.23 μg/m³. Furthermore, experiments conducted in areas with sparse ground monitoring stations demonstrate that the model maintains high predictive accuracy, further validating its applicability and generalization capability in data-limited scenarios. Moreover, the VDMS model adopts a modular design, offering strong scalability that allows its architecture to be adjusted according to specific requirements. This adaptability renders it suitable for monitoring various atmospheric pollutants, providing essential technical support for precise environmental management and air quality forecasting. Full article

In the context of cultural tourism integration, tourists’ spatial perception intention is an important carrier of spatial evaluation. In historic cultural districts represented by Jinan Mingfu City, tourists’ perceptual depth remains underexplored, leading to a misalignment between cultural tourism development and spatial quality needs. Taking Jinan Mingfu City as a representative case of a historic cultural district, while the living heritage model has revitalized local economies, the absence of a tourist perspective has resulted in misalignment between cultural tourism development and spatial quality requirements. This study establishes a technical framework encompassing “data crawling-factor aggregation-human-machine collaborative optimization”. It integrates Python web crawlers, SnowNLP sentiment analysis, and TF-IDF text mining technologies to extract physical elements; constructs a three-dimensional evaluation framework of “visual perception-spatial comfort-cultural experience” through SPSS principal component analysis; and quantifies physical element indicators such as green vision rate and signboard clutter index through street view semantic segmentation (OneFormer framework). A synergistic mechanism of machine scoring and manual double-blind scoring is adopted for correlation analysis to determine the impact degree of indicators and optimization strategies. This study identified that indicators such as green vision rate, shading facility coverage, and street enclosure ratio significantly influence tourist evaluations, with a severe deficiency in cultural spaces. Accordingly, it proposes targeted strategies, including visual landscape optimization, facility layout adjustment, and cultural scenario implementation. By breaking away from traditional qualitative evaluation paradigms, this study provides data-based support for the spatial quality enhancement of historic districts, thereby enabling the transformation of these areas from experience-oriented protection to data-driven intelligent renewal and promoting the sustainable development of cultural tourism. Full article

As a crucial characteristic waterfowl breed, the egg-laying performance of Lion-Headed Geese serves as a core indicator for precision breeding. Under large-scale flat rearing and selection practices, high phenotypic similarity among individuals within the same pedigree coupled with traditional manual observation and existing automation systems relying on fixed nesting boxes or RFID tags has posed challenges in achieving accurate goose–egg matching in dynamic environments, leading to inefficient individual selection. To address this, this study proposes YOLO-Goose, an improved YOLOv8s-based method, which designs five high-contrast neck rings (DoubleBar, Circle, Dot, Fence, Cylindrical) as individual identifiers. The method constructs a lightweight model with a small-object detection layer, integrates the GhostNet backbone to reduce parameter count by 67.2%, and employs the GIoU loss function to optimize neck ring localization accuracy. Experimental results show that the model achieves an F1 score of 93.8% and mAP50 of 96.4% on the self-built dataset, representing increases of 10.1% and 5% compared to the original YOLOv8s, with a 27.1% reduction in computational load. The dynamic matching algorithm, incorporating spatiotemporal trajectories and egg positional data, achieves a 95% matching rate, a 94.7% matching accuracy, and a 5.3% mismatching rate. Through lightweight deployment using TensorRT, the inference speed is enhanced by 1.4 times compared to PyTorch-1.12.1, with detection results uploaded to a cloud database in real time. This solution overcomes the technical bottleneck of individual selection in flat rearing environments, providing an innovative computer-vision-based approach for precision breeding of pedigree Lion-Headed Geese and offering significant engineering value for advancing intelligent waterfowl breeding. Full article

Background: Orbital floor blowout fractures (OFBF) can have serious consequences for the patient. Selecting the right treatment method and materials is essential. Krenkel’s maxillary sinus implant has been used successfully for more than 40 years in clinical practice. The aim of this study was to evaluate the long-term outcome of this implant compared to polydioxanone (PDS) sheets. Material and methods: This retrospective study examined a cohort of 82 OFBF patients over a seven-year period. Clinical and geometric data were collected. Defect size, location, and the volume of the herniated tissue were measured from conventional computer tomography (CT) or cone beam computer tomography (CBCT) scans. The relationship between ophthalmologic rehabilitation and treatment modality was analyzed using logistic regression. Results: The study included 82 patients, 28% female and 72% male, with a median age of 45.2 years. Defect size and hernia volume correlated with preoperative ophthalmological symptoms. At follow-up, 14.8% in the implant group and 28.6% in the PDS group showed mild visual impairment, with no severe diplopia. Conclusions: Our results suggest this method is a reliable and effective solution for repairing OFBFs and ophthalmologic rehabilitation. However, further research in a clinical controlled trial is needed. Full article

In 3D virtual environments, effective information presentation relies on clear label layouts to annotate complex objects. As label density increases, structured and adaptive layouts become critical to prevent visual clutter and maintain clarity. However, existing label layout methods often overlook the issue of double vision between labels caused by binocular disparity. Additionally, discrete update strategies in 3D label layouts struggle to maintain layout quality during viewpoint changes due to their limited real-time adaptability. To address these issues, this paper presents a binocular 3D dynamic label layout method based on a continuous updating strategy within a force-based framework, generating real-time binocular layouts to minimize visual confusion. Computational evaluation and a user study demonstrate that the proposed method significantly reduces double vision, as quantified by a reduction in double vision degree of up to 15.74%, along with label overlap and leader line crossings. The results further indicate that our approach surpasses existing methods in clarity, aesthetics, adaptability, and user satisfaction, while reducing label reading time and alleviating users’ physical and effort demands, particularly in high-density label environments. Full article

In the production process of cross-cutting boards, real-time measurement of dimensions online has been a long-standing technical problem in the production field. Currently, the detection of board dimensions in the production field relies on manual observation based on workers’ operational experience or stopping the machine for measurement. This paper proposes a machine vision-based real-time online measurement system for dimensional measurements of cross-cutting units. A certain angle measurement model is established by using a face-array industrial camera, and a more accurate edge contour extraction is realized by deep learning. A novel edge intersection extraction algorithm based on line fitting and least squares method was proposed to accurately measure the length, width, diagonal lines of cross-cutting boards using four intersection coordinates. The measurement of 100 cross-cutting boards in the industrial production site shows that the proposed online measurement system for cross-cut board dimensions in this article has high accuracy, with a length perception error of ±50 mm, width of ±2 mm, and diagonal difference of ±5 mm, meeting the production requirements in industrial settings. The on-site shutdown measurement work was reduced, thereby doubling the production efficiency and saving two staff members. Full article

Background: Patients with myasthenia gravis (MG) are susceptible to fractures due to glucocorticoid (GC) use and disease-related functional impairment affecting activities of daily living (ADL). The Fracture Risk Assessment Tool (FRAX^®) estimates fracture probability but does not incorporate disease-specific functional status. We investigated whether combining FRAX^® with the Myasthenia Gravis Activities of Daily Living (MG-ADL) scale improves fracture risk stratification in MG patients. Methods: This single-center prospective cohort study followed 53 MG patients for 10 years (2012–2022) at Toho University Ohashi Medical Center, Japan. Patients were categorized into four groups based on baseline FRAX^® probability (calculated with bone mineral density [BMD]) and MG-ADL scores using median splits: high FRAX^®/high MG-ADL (HH), high FRAX^®/low MG-ADL (HL), low FRAX^®/high MG-ADL (LH), and low FRAX^®/low MG-ADL (LL). The primary outcome was incident major osteoporotic fracture (MOF). Results: Over 10 years, nine MOFs occurred: seven in the HH group (43.8%), two in the HL group (16.7%), and none in the LH or LL groups. Fracture-free survival differed significantly among the groups (log-rank p < 0.001), with the HH group exhibiting the lowest survival rate. Baseline characteristics, including age, disease duration, MG severity scores, BMD, and FRAX^® scores, differed significantly among groups. Specific MG-ADL items reflecting greater impairment (impairment of ability to arise from a chair, double vision, and ptosis) were significantly more pronounced in the HH group at baseline. Conclusions: Combining baseline FRAX^® scores with the MG-ADL assessment effectively stratifies long-term MOF risk in patients with MG. Individuals with both high FRAX^® and high MG-ADL represent a particularly high-risk subgroup. This dual-assessment approach may improve the identification of patients requiring targeted preventive interventions. Full article

Accurate tomato yield estimation and ripeness monitoring are critical for optimizing greenhouse management. While manual counting remains labor-intensive and error-prone, this study introduces a novel vision-based framework for automated tomato counting in standardized greenhouse environments. The proposed method integrates YOLOv8-based detection, depth filtering, and an inter-frame prediction algorithm to address key challenges such as background interference, occlusion, and double-counting. Our approach achieves 97.09% accuracy in tomato cluster detection, with mature and immature single fruit recognition accuracies of 92.03% and 91.79%, respectively. The multi-target tracking algorithm demonstrates a MOTA (Multiple Object Tracking Accuracy) of 0.954, outperforming conventional methods like YOLOv8 + DeepSORT. By fusing odometry data from an inspection robot, this lightweight solution enables real-time yield estimation and maturity classification, offering practical value for precision agriculture. Full article