Search Results (188)

Predicting road crash severity remains a major challenge in transportation safety research, requiring models that combine predictive accuracy, interpretability, and computational efficiency. This study introduces a Multi-Risk Index based on Copula Integration (MRI-Copula)—a hybrid framework that integrates Categorical Boosting (CatBoost) with SHapley Additive exPlanations (SHAP) and Vine Copula dependence modeling to assess and predict crash severity. The approach leverages CatBoost–SHAP to quantify the marginal contribution of each risk factor while maintaining model transparency and employs copula-based tail dependence to capture the joint escalation of risk under extreme crash conditions. Using a dataset of 877 police-reported crashes from Jeddah, Saudi Arabia, the framework constructs three interpretable sub-indices—Environmental Risk Index (ERI), Behavioural Risk Index (BRI), and Systemic Risk Index (SRI)—representing distinct domains of crash causation. These indices are combined through a convex weighting parameter (α), optimized via cross-validation (optimal α = 0.80), ensuring a balanced integration of predictive and dependence-based information. Comparative evaluation across multiple classifiers—CatBoost, Light Gradient Boosting Machine (LightGBM), Histogram-based Gradient Boosting (HistGB), and Logistic Regression—demonstrated the robustness of the framework. The CatBoost + MRI-Copula configuration achieved the highest predictive performance (AUC = 0.986; F1 = 0.904), while LightGBM and HistGB offered comparable accuracy (AUC ≈ 0.958; F1 ≈ 0.89) at a fraction of the computational time (≤1 s versus 32 s for CatBoost), highlighting a trade-off between analytical precision and scalability. Consequently, the MRI-Copula framework provides a transparent and theoretically grounded foundation for data-driven road safety management. It bridges predictive analytics and decision support offering a scalable, interpretable, and policy-relevant tool for proactive crash risk mitigation. Full article

This paper presents a multiscale histogram excess-distribution strategy addressing the structural limitations (i.e., insufficient dark-region restoration, block artifacts, ringing effects, color distortion, and saturation loss) of contrast-limited adaptive histogram equalization (CLAHE) and retinex-based image-contrast enhancement techniques. This method adjusts the ratio between the uniform and weighted distribution of the histogram excess based on the average tile brightness. At the coarsest scale, excess pixels are redistributed to histogram bins initially occupied by pixels, maximizing detail restoration in dark areas. For medium and fine scales, the contrast enhancement strength is adjusted according to tile brightness to preserve local luminance transitions. Scale-specific lookup tables are bilinearly interpolated and merged at the pixel level. Background restoration corrects unnatural tone compression by referencing the original image, ensuring visual consistency. A ratio-based chroma adjustment and color-restoration function compensate for saturation degradation in retinex-based approaches. An asymmetric Gaussian offset correction preserves structural information and expands the global dynamic range. The experimental results demonstrate that this method enhances local and global contrast while preserving fine details in low light and high brightness. Compared with various existing methods, this method reproduces more natural color with superior image enhancement. Full article

Background/Objectives: The clinical management of adolescent idiopathic scoliosis (AIS) is hindered by the inability to accurately predict curve progression. Although skeletal maturity and the initial Cobb angle are established predictors of progression, their combined predictive accuracy remains limited. This study aimed to develop a robust and interpretable artificial intelligence (AI) system using deep learning (DL) models to predict the progression of scoliosis using only standing frontal radiographs. Methods: We conducted a multicenter study involving 542 patients with AIS. After excluding 52 borderline progression cases (6–9° progression in the Cobb angle), 294 and 196 patients were assigned to progression (≥10° increase) and non-progression (≤5° increase) groups, respectively, considering a 2-year follow-up. Frontal whole spinal radiographs were preprocessed using histogram equalization and divided into two regions of interest (ROIs) (ROI 1, skull base–femoral head; ROI 2, C7–iliac crest). Six pretrained DL models, including convolutional neural networks (CNNs) and transformer-based models, were trained on the radiograph images. Gradient-weighted class activation mapping (Grad-CAM) was further performed for model interpretation. Results: Ensemble models outperformed individual ones, with the average ensemble model achieving area under the curve (AUC) values of 0.769 for ROI 1 and 0.755 for ROI 2. Grad-CAM revealed that the CNNs tended to focus on the local curve apex, whereas the transformer-based models demonstrated global attention across the spine, ribs, and pelvis. Models trained on ROI 2 performed comparably with respect to those using ROI 1, supporting the feasibility of image standardization without a loss of accuracy. Conclusions: This study establishes the clinical potential of transformer-based DL models for predicting the progression of scoliosis using only plain radiographs. Our multicenter approach, high AUC values, and interpretable architectures support the integration of AI into clinical decision-making for the early treatment of AIS. Full article

With the proliferation of image-capturing and display-enabled IoT devices, ensuring the authenticity and integrity of visual data has become increasingly critical, especially in light of emerging cybersecurity threats and powerful generative AI tools. One of the major challenges in such sensor-based systems is the ability to protect privacy while maintaining data usability. Reversible data hiding has attracted growing attention due to its reversibility and ease of implementation, making it a viable solution for secure image communication in IoT environments. In this paper, we propose reversible data hiding techniques tailored to the content characteristics of images. Our approach leverages subsampling and quadtree partitioning, combined with multi-stage prediction schemes, to generate a predicted image aligned with the original. Secret information is embedded by analyzing the difference histogram between the original and predicted images, and enhanced through multi-round rotation techniques and a multi-level embedding strategy to boost capacity. By employing both subsampling and quadtree decomposition, the embedding strategy dynamically adapts to the inherent characteristics of the input image. Furthermore, we investigate the trade-off between embedding capacity and marked image quality. Experimental results demonstrate improved embedding performance, high visual fidelity, and low implementation complexity, highlighting the method’s suitability for resource-constrained IoT applications. Full article

Background/Objectives: Accurate segmentation of brain tumors in Magnetic Resonance Imaging (MRI) scans is critical for diagnosis and treatment planning due to their life-threatening nature. This study aims to develop a robust and automated method capable of precisely delineating heterogeneous tumor regions while improving segmentation accuracy and generalization. Methods: We propose Attention Res-UNet (ARU-Net), a novel Deep Learning (DL) architecture integrating residual connections, Adaptive Channel Attention (ACA), and Dimensional-space Triplet Attention (DTA) modules. The encoding module efficiently extracts and refines relevant feature information by applying ACA to the lower layers of convolutional and residual blocks. The DTA is fixed to the upper layers of the decoding module, decoupling channel weights to better extract and fuse multi-scale features, enhancing both performance and efficiency. Input MRI images are pre-processed using Contrast Limited Adaptive Histogram Equalization (CLAHE) for contrast enhancement, denoising filters, and Linear Kuwahara filtering to preserve edges while smoothing homogeneous regions. The network is trained using categorical cross-entropy loss with the Adam optimizer on the BTMRII dataset, and comparative experiments are conducted against baseline U-Net, DenseNet121, and Xception models. Performance is evaluated using accuracy, precision, recall, F1-score, Dice Similarity Coefficient (DSC), and Intersection over Union (IoU) metrics. Results: Baseline U-Net showed significant performance gains after adding residual connections and ACA modules, with DSC improving by approximately 3.3%, accuracy by 3.2%, IoU by 7.7%, and F1-score by 3.3%. ARU-Net further enhanced segmentation performance, achieving 98.3% accuracy, 98.1% DSC, 96.3% IoU, and a superior F1-score, representing additional improvements of 1.1–2.0% over the U-Net + Residual + ACA variant. Visualizations confirmed smoother boundaries and more precise tumor contours across all six tumor classes, highlighting ARU-Net’s ability to capture heterogeneous tumor structures and fine structural details more effectively than both baseline U-Net and other conventional DL models. Conclusions: ARU-Net, combined with an effective pre-processing strategy, provides a highly reliable and precise solution for automated brain tumor segmentation. Its improvements across multiple evaluation metrics over U-Net and other conventional models highlight its potential for clinical application and contribute novel insights to medical image analysis research. Full article

Diabetic Retinopathy (DR) is a leading cause of vision impairment globally, underscoring the need for accurate and early diagnosis to prevent disease progression. Although fundus imaging serves as a cornerstone of Computer-Aided Diagnosis (CAD) systems, several challenges persist, including lesion scale variability, blurry morphological patterns, inter-class imbalance, limited labeled datasets, and computational inefficiencies. To address these issues, this study proposes an end-to-end diagnostic framework that integrates an enhanced preprocessing pipeline with a novel deep learning architecture, Hierarchical-Inception-Residual-Dense Network (HIRD-Net). The preprocessing stage combines Contrast Limited Adaptive Histogram Equalization (CLAHE) with Dilated Difference of Gaussian (D-DoG) filtering to improve image contrast and highlight fine-grained retinal structures. HIRD-Net features a hierarchical feature fusion stem alongside multiscale, multilevel inception-residual-dense blocks for robust representation learning. The Squeeze-and-Excitation Channel Attention (SECA) is introduced before each Global Average Pooling (GAP) layer to refine the Feature Maps (FMs). It further incorporates four GAP layers for multi-scale semantic aggregation, employs the Hard-Swish activation to enhance gradient flow, and utilizes the Focal Loss function to mitigate class imbalance issues. Experimental results on the IDRiD-APTOS2019, DDR, and EyePACS datasets demonstrate that the proposed framework achieves 93.46%, 82.45% and 79.94% overall classification accuracy using only 4.8 million parameters, highlighting its strong generalization capability and computational efficiency. Furthermore, to ensure transparent predictions, an Explainable AI (XAI) approach known as Gradient-weighted Class Activation Mapping (Grad-CAM) is employed to visualize HIRD-Net’s decision-making process. Full article

This study investigates how phospholipid headgroups influence passive membrane penetration and structural impact of four nitrogen-, sulfur-, and oxygen-containing heterocycles (NSO-HETs)—N-methyl-2-pyrrolidone (PIR), 1,4-dioxane (DIOX), oxane (OXA), and phenol (PHE). Using all-atom molecular dynamics simulations combined with Accelerated Weight Histogram free energy calculations, the passive transport of NSO-HETs across DPPC, DPPE, DPPA, and DPPG bilayers was characterized. DPPG showed the highest membrane affinity, increasing permeability (logP_memb/bulk) by 27–64% compared to DPPE, associated with the lowest permeability and tightest lipid packing. Free energy barriers are also decreased in DPPG relative to DPPE; PIR’s central barrier dropped from 19.2 kJ/mol (DPPE) to 16.6 kJ/mol (DPPG), while DIOX’s barrier decreased from 7.2 to 5.2 kJ/mol. OXA exhibited the lowest central barriers (1.2–2.2 kJ/mol) and uniquely accumulated at higher concentrations in the bilayer center than in bulk water, with free energy ranging from −3.4 to −5.9 kJ/mol. PHE and OXA caused significant bilayer thinning (up to 11%) and reduced lipid tail order, especially in DPPE and DPPA. Concentration effects were most pronounced in DPPE, where high solute loading disrupted lipid order and altered free energy profiles. These results highlight the crucial role of headgroup identity in modulating NSO-HET membrane permeability and structural changes. Full article

Breast cancer remains one of the leading causes of mortality among women worldwide, highlighting the critical need for early and accurate detection methods. Traditional mammography, although widely used, has limitations, including radiation exposure and challenges in detecting early-stage lesions. Electrical Impedance Mammography (EIM) has emerged as a non-invasive and radiation-free alternative that assesses the density and electrical conductivity of breast tissue. EIM images consist of seven layers, each representing different tissue depths, offering a detailed representation of the breast structure. However, analyzing these layers individually can be redundant and complex, making it difficult to identify relevant features for lesion classification. To address this issue, advanced computational techniques are employed for image integration, such as the Root Mean Square ($C_{\mathrm{RMS}}$) Contrast and Contrast-Limited Adaptive Histogram Equalization (CLAHE), combined with the Coefficient of Variation (CV), CLAHE-based fusion, weighted average fusion, Gaussian pyramid fusion, and Wavelet–PCA fusion. Each method enhances the representation of tissue features, optimizing the image quality and diagnostic utility. This study evaluated the impact of these integration techniques on EIM image analysis, aiming to improve the accuracy and reliability of computational diagnostic models for breast cancer detection. According to the obtained results, the best performance was achieved using Wavelet–PCA fusion in combination with XGBoost as a classifier, yielding an accuracy rate of 89.5% and an F1-score of 81.5%. These results are highly encouraging for the further investigation of this topic. Full article

Leaf shape is a crucial visual cue for plant recognition. However, distinguishing among plants with high inter-class shape similarity remains a significant challenge, especially among cultivars within the same species where shape differences can be extremely subtle. To address this issue, we propose a novel shape representation and an advanced heterogeneous fusion framework for accurate leaf image retrieval. Specifically, based on the local polar coordinate system, multiscale analysis, and statistical histograms, we first propose local polar coordinate feature representation (LPCFR), which captures spatial distribution from two orthogonal directions while encoding local curvature characteristics. Next, we present heterogeneous feature fusion with exponential weighting and Ranking (HFER), which enhances the compatibility and robustness of fused features by applying exponential weighted normalization and ranking-based encoding within neighborhood distance measures. Extensive experiments on both species-level and cultivar-level leaf datasets demonstrate that the proposed representation effectively captures shape features, and the fusion framework successfully integrates heterogeneous features, outperforming state-of-the-art (SOTA) methods. Full article

Background/Objectives: Inflammatory bowel disease (IBD), encompassing Crohn’s disease and ulcerative colitis, is characterized by cycles of inflammation and tissue remodeling that can culminate in fibrosis. Differentiating between early inflammatory and fibrotic bowel wall changes remains a diagnostic challenge due to overlapping imaging features. This study aimed to assess the potential of elastography, specifically pixel histogram analysis, as a non-invasive method to identify acute inflammatory changes in a rat model of 2,4,6-trinitrobenzenesulfonic (TNBS)-induced colitis. Methods: Female CRL:Wi rats were randomized into control and experimental groups, with the latter receiving intracolonic TNBS to induce acute colitis. On day 7 post-induction, all animals underwent ultrasonographic and strain elastographic assessment of the distal colon using a standardized protocol. Histogram-based analysis of red, green, and blue pixel distributions was performed on elastographic video frames. Results were compared with histologic grading of inflammation and fibrosis using hematoxylin-eosin and Masson’s trichrome staining. Results: Rats with TNBS-induced colitis exhibited significant weight loss, increased bowel wall thickness (31.5% vs. controls, p < 0.01), and elevated elastographic pixel intensity across all color channels (p < 0.05). Histologically, experimental animals showed severe inflammation and early submucosal fibrosis. A strong positive correlation was found between elastographic histogram values and histologic fibrosis scores (r = 0.86, p < 0.01), confirming the technique’s diagnostic relevance. Conclusions: Elastographic pixel histogram analysis is a reproducible, non-invasive approach capable of distinguishing acute inflammatory changes and early fibrotic remodeling in experimental colitis. These findings support its potential application as a diagnostic adjunct in the early assessment and monitoring of IBD-related bowel wall changes. Full article

The Internet of Vehicles plays a crucial role in advancing intelligent transportation systems, with In-Vehicle Ethernet serving as the fundamental backbone network of the new generation of in-vehicle communication. However, In-Vehicle Ethernet faces various network security threats, including data theft, data tampering, and malicious attacks. This study focuses on network intrusion and security issues in In-Vehicle Ethernet, by analyzing the data characteristics of Audio Video Transport Protocol and potential network attack means. We innovatively propose a network intrusion detection method based on a weighted histogram algorithm. This method aims to enhance the security of In-Vehicle Ethernet. Experimental results show that the anomaly detection rate of the proposed weighted histogram algorithm in this study is 99.7%, which shows an improvement of 15.8% compared with the traditional Bayesian algorithm, and 6.9% higher than the decision tree algorithm. Thus, our approach enhances the stability and anti-attack ability of In-Vehicle Ethernet, providing a solid network security for In-Vehicle Networks. Full article

Background/Objectives: This work prospectively evaluates the vendor-provided Low Variance (LOVA) apparent diffusion coefficient (ADC) gradient nonlinearity correction (GNC) technique for primary tumors, neck nodal metastases, and normal masseter muscles in patients with head and neck cancers (HNCs). Methods: Multiple b-value diffusion-weighted (DW)-MR images were acquired on a 3.0 T scanner using a single-shot echo planar imaging (SS-EPI) and multi-shot (MS)-EPI for diffusion phantom materials (20% and 40% polyvinylpyrrolidone (PVP) in water). Pretreatment DW-MRI acquisitions were performed for sixty HNC patients (n = 60) who underwent chemoradiation therapy. ADC values with and without GNC were calculated offline using a monoexponential diffusion model over all b-values, relative percentage (r%) changes (Δ) in ADC values with and without GNC were calculated, and the ADC histograms were analyzed. Results: Mean ADC values calculated using SS-EPI DW data with and without GNC differed by ≤1% for both PVP20% and PVP40% at the isocenter, whereas off-center differences were ≤19.6% for both concentrations. A similar trend was observed for these materials with MS-EPI. In patients, the mean rΔADC (%) values measured with SS-EPI differed by 4.77%, 3.98%, and 5.68% for primary tumors, metastatic nodes, and masseter muscle. MS-EPI exhibited a similar result with 5.56%, 3.95%, and 4.85%, respectively. Conclusions: This study showed that the GNC method improves the robustness of the ADC measurement, enhancing its value as a quantitative imaging biomarker used in HNC clinical trials. Full article

This paper introduces a model that automates the diagnosis of a patient’s condition, reducing reliance on highly trained professionals, particularly in resource-constrained settings. To ensure data consistency, the dataset was preprocessed for uniformity in size, format, and color channels. Image quality was further enhanced using histogram equalization to improve the dynamic range. Lung regions were isolated using segmentation techniques, which also eliminated extraneous areas from the images. A modified segmentation-based cropping technique was employed to define an optimal cropping rectangle. Feature extraction was performed using persistent homology, deep learning, and hybrid methodologies. Persistent homology captured topological features across multiple scales, while the deep learning model leveraged convolutional transition equivariance, input-adaptive weighting, and the global receptive field provided by Vision Transformers. By integrating features from both methods, the classification model effectively predicted severity levels (mild, moderate, severe). The segmentation-based cropping method showed a modest improvement, achieving 80% accuracy, while stand-alone persistent homology features reached 66% accuracy. Notably, the hybrid model outperformed existing approaches, including SVM, ResNet50, and VGG16, achieving an accuracy of 82%. Full article

Background: The discovery of novel antibody–drug conjugates for low-expression human epidermal growth factor receptor 2 (HER2-low) breast cancer highlights the inadequacy of the conventional binary classification of HER2 status as either negative or positive. Identification of HER2-low breast cancer is crucial for selecting patients who may benefit from targeted therapies. This study aims to determine whether qualitative and quantitative magnetic resonance imaging (MRI) features can effectively reflect low-HER2-expression breast cancer. Methods: Pre-treatment breast MRI images from 232 patients with pathologically confirmed breast cancer were retrospectively analyzed. Both clinicopathologic and MRI features were recorded. Qualitative MRI features included Breast Imaging Reporting and Data System (BI-RADS) descriptors from dynamic contrast-enhanced MRI (DCE-MRI), as well as intratumoral T2 hyperintensity and peritumoral edema observed in T2-weighted imaging (T2WI). Quantitative features were derived from diffusion kurtosis imaging (DKI) using multiple b-values and included statistics such as mean, median, 5th and 95th percentiles, skewness, kurtosis, and entropy from apparent diffusion coefficient (ADC), D_app, and K_app histograms. Differences in clinicopathologic, qualitative, and quantitative MRI features were compared across groups, with multivariable logistic regression used to identify significant independent predictors of HER2-low breast cancer. The discriminative power of MRI features was assessed using receiver operating characteristic (ROC) curves. Results: HER2 status was categorized as HER2-zero (n = 60), HER2-low (n = 91), and HER2-overexpressed (n = 81). Clinically, estrogen receptor (ER), progesterone receptor (PR), hormone receptor (HR), and Ki-67 levels significantly differed between the HER2-low group and others (all p < 0.001). In MRI analyses, intratumoral T2 hyperintensity was more prevalent in HER2-low cases (p = 0.009, p = 0.008). Mass lesions were more common in the HER2-zero group than in the HER2-low group (p = 0.038), and mass shape (p < 0.001) and margin (p < 0.001) significantly varied between the HER2 groups, with mass shape emerging as an independent predictive factor (HER2-low vs. HER2-zero: p = 0.010, HER2-low vs. HER2-over: p = 0.012). Qualitative MRI features demonstrated an area under the curve (AUC) of 0.763 (95% confidence interval [CI]: 0.667–0.859) for distinguishing HER2-low from HER2-zero status. Quantitative features showed distinct differences between HER2-low and HER2-overexpression groups, particularly in non-mass enhancement (NME) lesions. Combined variables achieved the highest predictive accuracy for HER2-low status, with an AUC of 0.802 (95% CI: 0.701–0.903). Conclusions: Qualitative and quantitative MRI features offer valuable insights into low-HER2-expression breast cancer. While qualitative features are more effective for mass lesions, quantitative features are more suitable for NME lesions. These findings provide a more accessible and cost-effective approach to noninvasively identifying patients who may benefit from targeted therapy. Full article

The computational investigation of nonlinear mathematical models presents significant challenges due to their complex dynamics. This paper presents a computational study of a nonlinear hepatitis C virus model that accounts for the influence of alcohol consumption on disease progression. We employ periodic neural networks, optimized using a hybrid genetic algorithm and the interior-point algorithm, to solve a system of six coupled nonlinear differential equations representing hepatitis C virus dynamics. This model has not previously been solved using the proposed technique, marking a novel approach. The proposed method’s performance is evaluated by comparing the numerical solutions with those obtained from traditional numerical methods. Statistical measures such as mean absolute error, root mean square error, and Theil’s inequality coefficient are used to assess the accuracy and reliability of the proposed approach. The weight vector distributions illustrate how the network adapts to capture the complex nonlinear behavior of the disease. A comparative analysis with established numerical methods is provided, where performance metrics are illustrated using a range of graphical tools, including box plots, histograms, and loss curves. The absolute error values, ranging approximately from ${10}^{-6}$ to ${10}^{-10}$, demonstrate the precision, convergence, and robustness of the proposed approach, highlighting its potential applicability to other nonlinear epidemiological models. Full article