Search Results (1,030)

Water treadmill (WT) exercise is used for horses’ rehabilitation and training. Given that each training needs to be individualized for each horse, the goal is to assess whether infrared thermography (IRT) can serve as a non-invasive tool for daily monitoring of individual training and rehabilitation progress in horses undergoing WT exercise. Fifteen Polish Warmblood school horses were subjected to five WT sessions: dry treadmill, fetlock-depth water, fetlock-depth water with artificial river (AR), carpal-depth water, and carpal-depth water with AR. IRT images, collected pre- and post-exercise, were analyzed for the mean temperature (Tmean) and maximal temperature (Tmax) across 14 regions of interest (ROIs) representing the body surface overlying specific superficial muscles. While on a dry treadmill, Tmean and Tmax increased post-exercise in all ROIs; wetting of the hair coat limited surface temperature analysis in ROIs annotated on limbs. Tmax over the m. brachiocephalicus, m. trapezius pars cervicalis, m. triceps brachii, and m. semitendinosus increased during walking in carpal-depth water, which therefore may be suggested as an indirect indicator of increased activity related to forelimb protraction and flexion–extension of the limb joints. Tmax over the m. latissimus dorsi and m. longissimus increased during carpal-depth WT exercise with active AR mode, which may be suggested as an indicator of increased workload including vertical displacement of the trunk. Full article

Network connectivity and area coverage are the most important aspects in the applications of wireless sensor networks (WSNs). The resource and energy constraints of sensor nodes, operational conditions, and network size pose challenges to the optimal coverage of targets in the region of interest (ROI). The main idea is to achieve maximum area coverage and connectivity with strategic deployment and the minimal number of sensor nodes. This work addresses the problem of network area coverage in randomly distributed WSNs and provides an efficient deployment strategy using an enhanced version of cuckoo search optimization (ECSO). The “sequential update evaluation” mechanism is used to mitigate the dependency among dimensions and provide highly accurate solutions, particularly during the local search phase. During the preference random walk phase of conventional CSO, particle swarm optimization (PSO) with adaptive inertia weights is defined to accelerate the local search capabilities. The “opposition-based learning (OBL)” strategy is applied to ensure high-quality initial solutions that help to enhance the balance between exploration and exploitation. By considering the opposite of current solutions to expand the search space, we achieve higher convergence speed and population diversity. The performance of ECSO-OBL is evaluated using eight benchmark functions, and the results of three cases are compared with the existing methods. The proposed method enhances network coverage with a non-uniform distribution of sensor nodes and attempts to cover the whole ROI with a minimal number of sensor nodes. In a WSN with a 100 m² area, we achieved a maximum coverage rate of 98.45% and algorithm convergence in 143 iterations, and the execution time was limited to 2.85 s. The simulation results of various cases prove the higher efficiency of the ECSO-OBL method in terms of network coverage and connectivity in WSNs compared with existing state-of-the-art works. Full article

Background/Objectives: There is significant interest in using texture features to extract hidden image-based information. In medical imaging applications using radiomics, AI, or personalized medicine, the quest is to extract patient or disease specific information while being insensitive to other system or processing variables. While we use digital breast tomosynthesis (DBT) to show these effects, our results would be generally applicable to a wider range of other imaging modalities and applications. Methods: We examine factors in texture estimation methods, such as quantization, pixel distance offset, and region of interest (ROI) size, that influence the magnitudes of these readily computable and widely used image texture features (specifically Haralick’s gray level co-occurrence matrix (GLCM) textural features). Results: Our results indicate that quantization is the most influential of these parameters, as it controls the size of the GLCM and range of values. We propose a new multi-resolution normalization (by either fixing ROI size or pixel offset) that can significantly reduce quantization magnitude disparities. We show reduction in mean differences in feature values by orders of magnitude; for example, reducing it to 7.34% between quantizations of 8–128, while preserving trends. Conclusions: When combining images from multiple vendors in a common analysis, large variations in texture magnitudes can arise due to differences in post-processing methods like filters. We show that significant changes in GLCM magnitude variations may arise simply due to the filter type or strength. These trends can also vary based on estimation variables (like offset distance or ROI) that can further complicate analysis and robustness. We show pathways to reduce sensitivity to such variations due to estimation methods while increasing the desired sensitivity to patient-specific information such as breast density. Finally, we show that our results obtained from simulated DBT images are consistent with what we see when applied to clinical DBT images. Full article

Background: Choroidal nevi are common, benign tumors. These tumors rarely cause adverse retinal sequalae, but when they do, they can lead to disruption of the outer retina and vision loss. In this paper, we used high-resolution retinal imaging modalities, optical coherence tomography (OCT) and adaptive optics scanning laser ophthalmoscopy (AOSLO), to longitudinally monitor retinal sequelae of a submacular choroidal nevus. Methods: A 31-year-old female with a high-risk choroidal nevus resulting in subretinal fluid (SRF) and a 30-year-old control subject were longitudinally imaged with AOSLO and OCT in this study over 18 and 22 months. Regions of interest (ROI) including the macular region (where SRF was present) and the site of laser photocoagulation were imaged repeatedly over time. The depth of SRF in a discrete ROI was quantified with OCT and AOSLO images were assessed for visualization of photoreceptors and retinal pigmented epithelium (RPE). Cell-like structures that infiltrated the site of laser photocoagulation were measured and their count was assessed over time. In the control subject, images were assessed for RPE visualization and the presence and stability of cell-like structures. Results: We demonstrate that AOSLO can be used to assess cellular-level changes at small ROIs in the retina over time. We show the response of the retina to SRF and laser photocoagulation. We demonstrate that the RPE can be visualized when SRF is present, which does not appear to depend on the height of retinal elevation. We also demonstrate that cell-like structures, presumably immune cells, are present within and adjacent to areas of SRF on both OCT and AOSLO, and that similar cell-like structures infiltrate areas of retinal laser photocoagulation. Conclusions: Our study demonstrates that dynamic, cellular-level retinal responses to SRF and laser photocoagulation can be monitored over time with AOSLO in living humans. Many retinal conditions exhibit similar retinal findings and laser photocoagulation is also indicated in numerous retinal conditions. AOSLO imaging may provide future opportunities to better understand the clinical implications of such responses in vivo. Full article

Interpreting diagnostic imaging and identifying clinically relevant features remain challenging tasks, particularly for novice radiologists who often lack structured guidance and expert feedback. To bridge this gap, we propose an Eye-Gaze Guided Multimodal Fusion framework that leverages expert eye-tracking data to enhance learning and decision-making in medical image interpretation. By integrating chest X-ray (CXR) images with expert fixation maps, our approach captures radiologists’ visual attention patterns and highlights regions of interest (ROIs) critical for accurate diagnosis. The fusion model utilizes a shared backbone architecture to jointly process image and gaze modalities, thereby minimizing the impact of noise in fixation data. We validate the system’s interpretability using Gradient-weighted Class Activation Mapping (Grad-CAM) and assess both classification performance and explanation alignment with expert annotations. Comprehensive evaluations, including robustness under gaze noise and expert clinical review, demonstrate the framework’s effectiveness in improving model reliability and interpretability. This work offers a promising pathway toward intelligent, human-centered AI systems that support both diagnostic accuracy and medical training. Full article

3D object detection is a fundamental task in autonomous driving. In recent years, voxel-based methods have demonstrated significant advantages in reducing computational complexity and memory consumption when processing large-scale point cloud data. A representative method, Voxel-RCNN, introduces Region of Interest (RoI) pooling on voxel features, successfully bridging the gap between voxel and point cloud representations for enhanced 3D object detection. However, its robustness deteriorates when detecting distant objects or in the presence of noisy points (e.g., traffic signs and trees). To address this limitation, we propose an enhanced approach named Self-Attention Voxel-RCNN (SA-VoxelRCNN). Our method integrates two complementary attention mechanisms into the feature extraction phase. First, a full self-attention (FSA) module improves global context modeling across all voxel features. Second, a deformable self-attention (DSA) module enables adaptive sampling of representative feature subsets at strategically selected positions. After extracting contextual features through attention mechanisms, these features are fused with spatial features from the base algorithm to form enhanced feature representations, which are subsequently input into the region proposal network (RPN) to generate high-quality 3D bounding boxes. Experimental results on the KITTI test set demonstrate that SA-VoxelRCNN achieves consistent improvements in challenging scenarios, with gains of 2.49 and 1.87 percentage points at Moderate and Hard difficulty levels, respectively, while maintaining real-time performance at 22.3 FPS. This approach effectively balances local geometric details with global contextual information, providing a robust detection solution for autonomous driving applications. Full article

Barcodes are ubiquitous in manufacturing and logistics, but defects can reduce decoding efficiency and disrupt the supply chain. Existing studies primarily focus on a single barcode type or rely on small-scale datasets, limiting generalizability. We propose Y8-LiBAR Net, a lightweight two-stage framework for multi-type barcode defect detection. In stage 1, a YOLOv8n backbone localizes 1D and 2D barcodes in real time. In stage 2, a dual-branch network integrating ResNet50 and ViT-B/16 via hierarchical attention performs three-class classification on cropped regions of interest (ROIs): intact, defective, and non-barcode. Experiments conducted on the public BarBeR dataset, covering planar/non-planar surfaces, varying illumination, and sensor noise, show that Y8-LiBARNet achieves a detection-stage mAP@0.5 = 0.984 (1D: 0.992; 2D: 0.977) with a peak F1 score of 0.970. Subsequent defect classification attains 0.925 accuracy, 0.925 recall, and a 0.919 F1 score. Compared with single-branch baselines, our framework improves overall accuracy by 1.8–3.4% and enhances defective barcode recall by 8.9%. A Cohen’s kappa of 0.920 indicates strong label consistency and model robustness. These results demonstrate that Y8-LiBARNet delivers high-precision real-time performance, providing a practical solution for industrial barcode quality inspection. Full article

Background: This study aimed to develop a nomogram based on the most relevant clinical, CT, and radiomic features comprising 11 key signatures (2 clinical, 2 CT-based, and 7 radiomic) for the non-invasive prediction of the EGFR mutation status and to support the timely initiation of tyrosine kinase inhibitor (TKI) therapy in patients with non-small cell lung cancer (NSCLC) adenocarcinoma. Methods: Retrospective real-world data were collected from 521 patients with histologically confirmed NSCLC adenocarcinoma who underwent CT imaging and either surgical resection or pathological biopsy for EGFR mutation testing. Five Random Forest classification models were developed and trained on various datasets constructed by combining clinical, CT, and radiomic features extracted from CT image regions of interest (ROIs), with and without feature preselection. Results: The model trained exclusively on the most relevant clinical, CT, and radiomic features demonstrated superior predictive performance compared to the other models, with strong discrimination between EGFR-mutant and wild-type cases (AUC = 0.88; macro-average = 0.90; micro-average = 0.89; precision = 0.90; recall = 0.94; F1-score = 0.91; and accuracy = 0.87). Conclusions: A nomogram constructed using a Random Forest model trained solely on the most informative clinical, CT, and radiomic features outperformed alternative approaches in the non-invasive prediction of the EGFR mutation status, offering a promising decision-support tool for precision treatment planning in NSCLC. Full article

Buildings are a significant component of urban space and are essential to smart cities, catastrophe monitoring, and land use planning. However, precisely extracting building polygons from remote sensing images remains difficult because of the variety of building designs and intricate backgrounds. This paper proposes an end-to-end polygon dynamic adjustment algorithm (PDAA) to improve the accuracy and geometric consistency of building contour extraction by dynamically generating and optimizing polygon vertices. The method first locates building instances through the region of interest (RoI) to generate initial polygons, and then uses four core modules for collaborative optimization: (1) the feature enhancement module captures local detail features to improve the robustness of vertex positioning; (2) the contour vertex tuning module fine-tunes vertex coordinates through displacement prediction to enhance geometric accuracy; (3) the learnable redundant vertex removal module screens key vertices based on a classification mechanism to eliminate redundancy; and (4) the missing vertex completion module iteratively restores missed vertices to ensure the integrity of complex contours. PDAA dynamically adjusts the number of vertices to adapt to the geometric characteristics of different buildings, while simplifying the prediction process and reducing computational complexity. Experiments on public datasets such as WHU, Vaihingen, and Inria show that PDAA significantly outperforms existing methods in terms of average precision (AP) and polygon similarity (PolySim). It is at least 2% higher than existing methods in terms of average precision (AP), and the generated polygonal contours are closer to the real building geometry. Values of 75.4% AP and 84.9% PolySim were achieved on the WHU dataset, effectively solving the problems of redundant vertices and contour smoothing, and providing high-precision building vector data support for scenarios such as smart cities and emergency response. Full article

Background/Objectives: The glymphatic system plays a crucial role in clearing brain metabolic waste, and its dysfunction has been correlated to various neurological disorders. The Diffusion Tensor Imaging Analysis Along the Perivascular Space (DTI-ALPS) index has been proposed as a non-invasive marker of glymphatic function by measuring diffusivity along perivascular spaces; however, its sensitivity to sleep-related changes in glymphatic activity has not yet been validated. This study aimed to evaluate the feasibility of using the DTI-ALPS index as a quantitative marker of dynamic glymphatic activity during sleep. Methods: Diffusion tensor imaging (DTI) data were obtained from 12 healthy male participants (age = 24.44 ± 2.5 years; Pittsburgh Sleep Quality Index (PSQI) < 5), once while awake and 16 times during sleep, following 24 h sleep deprivation and administration of 10 mg zolpidem. Simultaneous MR-compatible electroencephalography was used to determine whether the subject was asleep or awake. DTI preprocessing included eddy current correction and tensor fitting. The DTI-ALPS index was calculated from nine regions of interest in projection and association areas aligned to standard space. The final analysis included nine participants (age = 24.56 ± 2.74 years; PSQI < 5) who maintained a continuous sleep state for 1 h without awakening. Results: Among nine ROI pairs, three showed significant increases in the DTI-ALPS index during sleep compared to wakefulness (Friedman test; p = 0.027, 0.029, 0.034). These ROIs showed changes at 14, 19, and 25 min after sleep induction, with FDR-corrected p-values of 0.024, 0.018, and 0.018, respectively. Conclusions: This study demonstrated a statistically significant increase in the DTI-ALPS index within 30 min after sleep induction through time-series DTI analysis during wakefulness and sleep, supporting its potential as a biomarker reflecting glymphatic activity. Full article

Beef fat injection technology, used to enhance the perceived quality of lower-grade meat, often results in artificially marbled beef that mimics the visual traits of Wagyu, characterized by dense fat distribution. This practice, driven by the high cost of Wagyu and the affordability of fat-injected beef, has led to the proliferation of mislabeled “Wagyu-grade” products sold at premium prices, posing potential food safety risks such as allergen exposure or consumption of unverified additives, which can adversely affect consumer health. Addressing this, this study introduces a smart sensing system integrated with handheld mobile devices, enabling consumers to capture beef images during purchase for real-time health-focused assessment. The system analyzes surface texture and color, transmitting data to a server for classification to determine if the beef is artificially marbled, thus supporting informed dietary choices and reducing health risks. Images are processed by applying a region of interest (ROI) mask to remove background noise, followed by partitioning into grid blocks. Local binary pattern (LBP) texture features and RGB color features are extracted from these blocks to characterize surface properties of three beef types (Wagyu, regular, and fat-injected). A support vector machine (SVM) model classifies the blocks, with the final image classification determined via majority voting. Experimental results reveal that the system achieves a recall rate of 95.00% for fat-injected beef, a misjudgment rate of 1.67% for non-fat-injected beef, a correct classification rate (CR) of 93.89%, and an F1-score of 95.80%, demonstrating its potential as a human-centered healthcare tool for ensuring food safety and transparency. Full article

Background: The ability of Graph Neural Networks (GNNs) to analyse brain structural patterns in various kinds of neurodegenerative diseases, including Parkinson’s disease (PD), has drawn a lot of interest recently. One emerging technique in this field is brain age prediction, which estimates biological age to identify ageing patterns that may serve as biomarkers for such disorders. However, a significant problem with most of the GNNs is their depth, which can lead to issues like oversmoothing and diminishing gradients. Methods: In this study, we propose SAGEFusionNet, a GNN architecture specifically designed to enhance brain age prediction and assess PD-related brain ageing patterns using T1-weighted structural MRI (sMRI). SAGEFusionNet learns important ROIs for brain age prediction by incorporating ROI-aware pooling at every layer to overcome the above challenges. Additionally, it incorporates multi-layer feature fusion to capture multi-scale structural information across the network hierarchy and auxiliary supervision to enhance gradient flow and feature learning at multiple depths. The dataset utilised in this study was sourced from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. It included a total of 580 T1-weighted sMRI scans from healthy individuals. The brain sMRI scans were parcellated into 56 regions of interest (ROIs) using the LPBA40 brain atlas in CAT12. The anatomical graph was constructed based on grey matter (GM) volume features. This graph served as input to the GNN models, along with GM and white matter (WM) volume as node features. All models were trained using 5-fold cross-validation to predict brain age and subsequently tested for performance evaluation. Results: The proposed framework achieved a mean absolute error (MAE) of $4.24\pm 0.38$ years and a mean Pearson’s Correlation Coefficient (PCC) of $0.72\pm 0.03$ during cross-validation. We also used 215 PD patient scans from the Parkinson’s Progression Markers Initiative (PPMI) database to assess the model’s performance and validate it. The initial findings revealed that out of 215 individuals with Parkinson’s disease, 213 showed higher and 2 showed lower predicted brain ages than their actual ages, with a mean MAE of 13.36 years (95% confidence interval: 12.51–14.28). Conclusions: These results suggest that brain age prediction using the proposed method may provide important insights into neurodegenerative diseases. Full article

Breast cancer is a leading cause of mortality among women worldwide. Early detection is crucial for increasing patient survival rates. Artificial intelligence, particularly convolutional neural networks (CNNs), has enabled the development of effective diagnostic systems by digitally processing mammograms. CNNs have been widely used for the classification of breast cancer in images, obtaining accurate results similar in many cases to those of medical specialists. This work presents a hybrid feature extraction approach for breast cancer detection that employs variants of EfficientNetV2 network and convenient image representation based on phase features. First, a region of interest (ROI) is extracted from the mammogram. Next, a three-channel image is created using the local phase, amplitude, and orientation features of the ROI. A feature vector is constructed for the processed mammogram using the developed CNN model. The size of the feature vector is reduced using simple statistics, achieving a redundancy suppression of $99.65\%$. The reduced feature vector is classified as either malignant or benign using a classifier ensemble. Experimental results using a training/testing ratio of 70/30 on 15,506 mammography images from three datasets produced an accuracy of $86.28\%$, a precision of $78.75\%$, a recall of $86.14\%$, and an F1-score of $80.09\%$ with the modified EfficientNetV2 model and stacking classifier. However, an accuracy of $93.47\%$, a precision of $87.61\%$, a recall of $93.19\%$, and an F1-score of $90.32\%$ were obtained using only CSAW-M dataset images. Full article

With the rapid expansion of Internet of Things (IoT) and edge computing applications, efficient video transmission under constrained bandwidth and limited computational resources has become increasingly critical. In such environments, perception-based video coding plays a vital role in maintaining acceptable visual quality while minimizing bit rate and processing overhead. Although newer video coding standards have emerged, H.264/AVC remains the dominant compression format in many deployed systems, particularly in commercial CCTV surveillance, due to its compatibility, stability, and widespread hardware support. Motivated by these practical demands, this paper proposes a perception-based video coding algorithm specifically tailored for low-bit-rate H.264/AVC applications. By targeting regions most relevant to the human visual system, the proposed method enhances perceptual quality while optimizing resource usage, making it particularly suitable for embedded systems and bandwidth-limited communication channels. In general, regions containing human faces and those exhibiting significant motion are of primary importance for human perception and should receive higher bit allocation to preserve visual quality. To this end, macroblocks (MBs) containing human faces are detected using the Viola–Jones algorithm, which leverages AdaBoost for feature selection and a cascade of classifiers for fast and accurate detection. This approach is favored over deep learning-based models due to its low computational complexity and real-time capability, making it ideal for latency- and resource-constrained IoT and edge environments. Motion-intensive macroblocks were identified by comparing their motion intensity against the average motion level of preceding reference frames. Based on these criteria, a dynamic quantization parameter (QP) adjustment strategy was applied to assign finer quantization to perceptually important regions of interest (ROIs) in low-bit-rate scenarios. The experimental results show that the proposed method achieves superior subjective visual quality and objective Peak Signal-to-Noise Ratio (PSNR) compared to the standard JM software and other state-of-the-art algorithms under the same bit rate constraints. Moreover, the approach introduces only a marginal increase in computational complexity, highlighting its efficiency. Overall, the proposed algorithm offers an effective balance between visual quality and computational performance, making it well suited for video transmission in bandwidth-constrained, resource-limited IoT and edge computing environments. Full article

Background/Objectives: Diagnosis of Persistent Spinal Pain Syndrome Type 2 (PSPS-T2) currently lacks objective biomarkers. Therefore, this retrospective study aimed to investigate differences in glucose metabolism in the axial musculoskeletal system in PSPS-T2 patients by means of [¹⁸F]FDG PET-CT imaging. Methods: Nine PSPS-T2 patients (five females, four males; mean age of 53 ± 4.82 years) and nine age- and gender-matched healthy controls (five females, four males; mean age of 53 ± 3.91 years) were included. For each participant, 24 regions of interest (ROIs) were manually drawn, including areas of the vertebral endplates, the intervertebral discs, and the psoas muscles. For each ROI, the mean standardized uptake values (SUVs) were assessed. Group differences were evaluated using repeated measures ANOVA with Bonferroni-adjusted post-hoc pairwise comparisons. Additionally, Pearson correlation analyses examined associations between SUV_mean values and the Numerical Rating Scale (NRS) pain scores. Results: Results demonstrated significantly higher SUV_mean values in healthy controls compared to PSPS-T2 patients, particularly at the superior endplates of L4 and S1, the intervertebral discs at L4-L5 and L5-S1, and the posterior endplates of L4 and L5. Although PSPS-T2 patients exhibited higher SUV_mean values than controls in the psoas muscle, these differences were not statistically significant. Additionally, no significant correlations were found between SUV_mean values and NRS pain scores, suggesting that metabolic activity alone does not directly reflect pain severity. Conclusions: Despite the limited sample size of this pilot study, the metabolic fingerprint of the axial musculoskeletal system was shown to be distinctly different in PSPS-T2 patients compared to healthy controls. This could lead to an improved understanding of PSPS-T2 pathophysiology and might open new doors for better diagnosis and treatment strategies. Full article