Search Results (374)

Aiming at the core problem of high false and missed alarm rate and insufficient interference resistance of existing smoke and fire detection algorithms in complex scenes, this paper proposes a target detection network based on improved feature pyramid structure. By constructing a Context Guided Convolutional Block instead of the traditional convolutional operation, the detected target and the surrounding environment information are fused with secondary features while reconfiguring the feature dimensions, which effectively solves the problem of edge feature loss in the down-sampling process. The Poly Kernel Inception Block is designed, and a multi-branch parallel network structure is adopted to realize multi-scale feature extraction of the detected target, and the collaborative characterization of the flame profile and smoke diffusion pattern is realized. In order to further enhance the logical location sensing ability of the target, a Manhattan Attention Mechanism Unit is introduced to accurately capture the spatial and temporal correlation characteristics of the flame and smoke by establishing a pixel-level long-range dependency model. Experimental tests are conducted using a self-constructed high-quality smoke and fire image dataset, and the results show that, compared with the existing typical lightweight smoke and fire detection models, the present algorithm has a significant advantage in detection accuracy, and it can satisfy the demand for real-time detection. 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

Diagnosing gastrointestinal disorders (GIDs) remains a significant challenge, particularly when relying on wireless capsule endoscopy (WCE), which lacks advanced imaging enhancements like Narrow Band Imaging (NBI). To address this, we propose a novel framework, the Spectrum-Aided Vision Enhancer (SAVE), especially designed to transform standard white light (WLI) endoscopic images into spectrally enriched representations that emulate both hyperspectral imaging (HSI) and NBI formats. By leveraging color calibration through the Macbeth Color Checker, gamma correction, CIE 1931 XYZ transformation, and principal component analysis (PCA), SAVE reconstructs detailed spectral information from conventional RGB inputs. Performance was evaluated using the Kvasir-v2 dataset, which includes 6490 annotated images spanning eight GI-related categories. Deep learning models like Inception-Net V3, MobileNetV2, MobileNetV3, and AlexNet were trained on both original WLI- and SAVE-enhanced images. Among these, MobileNetV2 achieved an F1-score of 96% for polyp classification using SAVE, and AlexNet saw a notable increase in average accuracy to 84% when applied to enhanced images. Image quality assessment showed high structural similarity (SSIM scores of 93.99% for Olympus endoscopy and 90.68% for WCE), confirming the fidelity of the spectral transformations. Overall, the SAVE framework offers a practical, software-based enhancement strategy that significantly improves diagnostic accuracy in GI imaging, with strong implications for low-cost, non-invasive diagnostics using capsule endoscopy systems. Full article

Background and Objectives: Osteoarthritis (OA) is a chronic, degenerative joint disease. The main symptoms include pain that can cause loss of function and stiffness, as well as swelling, reduced range of motion, crepitus, joint deformity, and muscle weakness. It leads to irreversible structural changes, that in advanced stages can require surgical interventions. The aim of this review was to summarize the current literature about the role of virtual reality (VR), exergames and digital technologies in patients with knee osteoarthritis before or after total knee arthroplasty, to understand if it is possible to prevent and reduce the symptoms and if these new technologies are more effective than conventional rehabilitation therapies. Materials and Methods: We conducted a systematic search of PubMed, Cochrane Library, Scopus, and PEDro from inception to November 2024. The review adhered to the PRISMA 2020 guidelines, and the protocol was prospectively registered in PROSPERO (registration number: CRD42024541890). We included randomized controlled trials (RCTs) enrolling participants aged 60 years or older, in which VR or telerehabilitation programs were compared with conventional rehabilitation approaches. Eligible studies had to report at least one of the following outcomes: pain, functionality, stability, or adherence. Two independent reviewers screened titles and abstracts, assessed full-text eligibility, extracted data, and evaluated the risk of bias using the Cochrane Risk of Bias 2 (RoB 2) tool. Results: Fourteen randomized controlled trails (RCTs) (1123 participants; mean age 68.2 years) were included. VR and telerehabilitation generally outperformed conventional rehabilitation for pain (8/13 studies, −0.9 to −2.3 VAS points) and functionality (7/13 studies, WOMAC improvement 8–15%, TUG −1.2 to −2.8 s). Compliance was higher in most technology-assisted programs (6/7 studies, 70–100% adherence). Stability outcomes were less consistent, with only 1/4 studies showing clear benefit. One study favored conventional rehabilitation for functionality. Overall risk of bias was low-to-moderate, with heterogeneity mainly driven by intervention duration, platform type, and supervision level. Conclusions: Structured telerehabilitation, non-immersive VR, and interactive online exercise programs, especially those offering real-time feedback, show comparable or superior benefits to conventional rehabilitation in older adults with knee OA or after TKA, particularly for pain reduction, functional gains, and adherence. These approaches enhance accessibility and home-based care, supporting their integration into clinical practice when in-person therapy is limited. Full article

In this paper, a wideband metasurface absorber is proposed by utilizing transfer learning and a conditional deep convolutional generative adversarial network (CDCGAN). This approach involves introducing a forward prediction neural network to predict the spectral curve of a metasurface absorber, as well as a generative adversarial network for the inverse design of a metasurface absorber. After comparing different pre-trained models, a transfer learning network (TLN) based on GoogleNet-InceptionV3 is incorporated into the design process to reduce the amount of training data required. Based on the pixelated metasurface with a common effect of metallic pixels and resistive film pixels, a broadband electromagnetic absorber was designed through the CDCGAN model. For the application target of the C-band, a pixelated broadband metasurface Absorber I has been designed, which can achieve an absorption effect of less than −8 dB in the range of 6.5–8 GHz, and the absorption performance reaches less than −15 dB near the resonant frequency point of 7 GHz. Further lightweight optimization design was carried out, and the metasurface Absorber II was designed for application in the X-band, which has an absorption bandwidth below −8 dB at 9–12 GHz. The reflectivity curve measured by the experiment is in good agreement with that of the simulation result. Of note, our methodology aims to reversely engineer suitable absorbing structures based on customer-defined spectrums, which may bear some significance to the rapid design of broadband metasurface absorbers. Full article

Background: There is a worldwide increase in the prevalence of type 2 diabetes, and strategies for managing this condition include dietary interventions. These interventions include the use of a low-glycaemic index diet, high-fibre and prebiotic diets, and low-carbohydrate diets (LCDs), which improve glycaemic control, reduce the risk of diabetic complications, and promote health. However, the definition of LCDs varies across the literature, and the use of LCDs in managing people with diabetes is often seen as controversial. Therefore, the aim of this review is to examine current advances and future prospects in the use of LCDs in managing people with type 2 diabetes. Method: A systematic review of randomised controlled trials, which applied both the PRISMA and PICOS frameworks. Databases including MEDLINE, APA PsycInfo, Academic Search Premier, CINAHL Plus with Full Text, APA PsycArticles, and Psychology and Behavioural Sciences Collection were searched through EBSCOHost. The EMBASE database and reference list of articles were also searched for articles of interest. Two researchers conducted the searches independently from database inception to 28 August 2025. However, based on the inclusion criteria, the year of publication of studies was restricted to articles published from 2021. The search terms were combined using Boolean operators (AND/OR), and duplicates were removed in EndNote. The articles were screened for eligibility based on inclusion and exclusion criteria by two researchers. Results: The findings identified that an LCD is significantly (p < 0.05) more effective in reducing glycaemic parameters compared to a usual diet, standard care, or a control diet in people with type 2 diabetes. Similarly, the effect of LCD was significant (p < 0.05) in reducing BMI in patients with type 2 diabetes compared with the control diet. However, an LCD did not appear to have a significant (p > 0.05) effect on lipid parameters compared to a control diet. Conclusion: This systematic review found that LCDs are significantly (p < 0.05) more effective in promoting glycaemic control than a usual diet, standard care, or a control diet in people with type 2 diabetes. In addition, LCDs can be an effective strategy for reducing BMI in individuals with type 2 diabetes, particularly when implemented as part of a structured, sustained dietary intervention. However, there was variability in the findings of the studies included with respect to glycaemic control and BMI. Furthermore, the impact of LCD on glycaemic control did not appear sustainable in the long term. LCDs did not have a significant (p > 0.05) effect on lipid parameters compared to a control diet. Full article

Although diagnostic criteria and research are constantly advancing, distinguishing between progressive supranuclear palsy (PSP) and corticobasal syndrome (CBS) remains a significant challenge. This difficulty stems from their similar clinical symptoms and the lack of reliable biomarkers. In this work, we present a detailed review of fluorodeoxyglucose (FDG)–positron emission tomography (PET), exploring its potential role in differentiating PSP and CBS, drawing on their established utility in other neurodegenerative diseases. We searched the PubMed database from its inception for original research articles assessing the utility of FDG-PET for the diagnosis or differential diagnosis of PSP and CBS from other neurodegenerative conditions. A total of 91 studies were eligible. These 91 studies were categorized as follows: (a) 20 studies included only patients with PSP, (b) 15 studies included only patients with CBS, (c) 39 studies involved patients with Parkinson’s disease and atypical Parkinsonian disorders, including subgroups of PSP and/or CBS, and (d) 17 studies compared patients with PSP and/or CBS to individuals with Alzheimer’s disease, frontotemporal dementia, or other dementias. Most FDG-PET studies involving PSP and CBS were not specifically designed for these disorders. An additional obstacle lies in the methodological variability across studies. Despite several studies achieving high diagnostic accuracy for PSP and/or CBS with specificity exceeding 90% using FDG-PET, sensitivity remains considerably lower. CBS appears to have a distinct hypometabolic pattern compared to PSP, marked by asymmetry and predominant cortical involvement. CBS more often affects posterior cortical regions (parietal and posterior parts of the frontal cortex, and sometimes temporal and occipital parts) and the thalamus, whereas PSP appears to affect the striatum, frontal cortex, anterior cingulate, and subtentorial structures, typically in a more symmetrical manner. Large, multicenter studies are needed, utilizing standardized imaging and protocols. Full article

To mitigate flood damage caused by overflow from a levee, it is essential to prevent the levee failure or extend the time to breaching. Although turfgrass on a levee slope is effective in suppressing erosion, insufficient maintenance can reduce its coverage. When overtopping occurs under such non-uniform turfgrass conditions, the flow tends to entrain air. In spillways, air entrainment is known to reduce friction loss; therefore, it may also contribute to lowering shear stress and erosion depth. This study conducted flume experiments with artificial turf arranged in various patterns on levee slopes to investigate flow patterns, air entrainment, and erosion. The flow pattern changed depending on the turf arrangement and overflow depth, and air entrainment occurred due to water surface fluctuations around the turfgrass. The inception point of air entrainment was found to be similar to or shorter than that observed in stepped spillways. Furthermore, the experiments showed a tendency for erosion depth to decrease once air entrainment is fully developed. This finding is significant because it suggests that erosion can potentially be minimized not only by reinforcing the levee structure itself but also by modifying flow characteristics through designs that promote air entrainment. Full article

Ulcerative colitis, polyps, esophagitis, and other gastrointestinal (GI) diseases significantly impact health, making early detection crucial for reducing mortality rates and improving patient outcomes. Traditional white light imaging (WLI) is commonly used during endoscopy to identify abnormalities in the gastrointestinal tract. However, insufficient contrast often limits its effectiveness, making it challenging to distinguish between healthy and unhealthy tissues, particularly when identifying subtle mucosal and vascular abnormalities. These limitations have prompted the need for more advanced imaging techniques that enhance pathological visualization and facilitate early diagnosis. Therefore, this study investigates the integration of the Spectrum-Aided Vision Enhancer (SAVE) mechanism to improve WLI images and increase disease detection accuracy. This approach transforms standard WLI images into hyperspectral imaging (HSI) representations, creating narrow-band imaging (NBI-like) visuals with enhanced contrast and tissue differentiation, thereby improving the visualization of vascular and mucosal structures critical for diagnosing GI disorders. This transformation allows for a clearer representation of blood vessels and membrane formations, which is essential for determining the presence of GI diseases. The dataset for this study comprises WLI images alongside SAVE-enhanced images, including four categories: ulcerative colitis, polyps, esophagitis, and healthy GI tissue. These images are organized into training, validation, and test sets to develop a deep learning-based classification model. Utilizing principal component analysis (PCA) and multiple regression analysis for spectral standardization ensures that the improved images retain spectral characteristics that are vital for clinical applications. By merging deep learning techniques with advanced imaging enhancements, this study aims to create an artificial intelligence (AI)–driven diagnostic system capable of early and accurate detection of GI diseases. InceptionV3 attained an overall accuracy of 94% in both scenarios; SAVE produced a modest enhancement in the ulcerative colitis F1-score from 92% to 93%, while the F1-scores for other classes exceeded 96%. SAVE resulted in a 10% increase in YOLOv8x accuracy, reaching 89%, with ulcerative colitis F1 improving to 82% and polyp F1 rising to 76%. VGG16 enhanced accuracy from 85% to 91%, and the F1-score for polyps improved from 68% to 81%. These findings confirm that SAVE enhancement consistently improves disease classification across diverse architectures, offers a practical, hardware-independent approach to hyperspectral-quality images, and enhances the accuracy of gastrointestinal screening. Furthermore, this research seeks to provide a practical and effective solution for clinical applications, improving diagnostic accuracy and facilitating superior patient care. Full article

The promise of achieving desired outcomes in community-engaged research relies upon an ongoing and long-term connection between the community and researchers. However, many community–researcher relationships begin and end in the confines of a single project, often precluding the sustainability and scalability of programs and initiatives that can benefit the community. Few examples exist in the literature, especially for the focus of this paper—school behavioral health (SBH)—to understand how the complex, challenging, and nuanced process of continued engagement between researchers and community members can be sustained and succeed. In this article, we chronicle the development of the Southeastern School Behavioral Health Community across 13 years, from its inception in a single state to its regional expansion through two research awards, to illustrate how long-term community engagement and a history of community connections can shape SBH research and practice across project action cycles. We describe the strengths, challenges, and lessons learned from this long-term community engagement experience. Numerous examples illustrate proactive and responsive strategies for initiating and sustaining community engagement throughout all phases of the longitudinal initiative and demonstrate tangible ways in which meaningful engagement influenced both research and practice. The reflections include the extent to which engagement principles of the research funder (the Patient-Centered Outcomes Research Institute, PCORI) were enacted during this research program; our roles as researchers, facilitators, and community members; the impact of the COVID-19 pandemic; engagement facilitators and structures; and what was achieved regarding levels of engagement. Future directions are provided for sustaining interconnected, community-engaged research and practice in SBH. Full article

Background: The limited availability of cardiac MRI data significantly constrains deep learning applications in cardiovascular imaging, necessitating innovative approaches to address data scarcity while preserving critical cardiac anatomical features. Methods: We developed a specialized denoising diffusion probabilistic model incorporating an attention-enhanced UNet architecture with strategically placed attention blocks across five hierarchical levels. The model was trained and evaluated on the OCMR dataset and compared against state-of-the-art generative approaches including StyleGAN2-ADA, WGAN-GP, and VAE baselines. Results: Our approach achieved superior image quality with a Fréchet Inception Distance of 77.78, significantly outperforming StyleGAN2-ADA (117.70), WGAN-GP (227.98), and VAE (325.26). Structural similarity metrics demonstrated excellent performance (SSIM: 0.720 ± 0.143; MS-SSIM: 0.925 ± 0.069). Clinical validation by cardiac radiologists yielded discrimination accuracy of only 60.0%, indicating near-realistic image quality that is challenging for experts to distinguish from real images. Comprehensive anatomical analysis revealed that 13 of 20 cardiac metrics showed no significant differences between real and synthetic images, with particularly strong preservation of left ventricular features. Discussion: The generated synthetic images demonstrate high anatomical fidelity with expert-level quality, as evidenced by the difficulty radiologists experienced in distinguishing synthetic from real images. The strong preservation of cardiac anatomical features, particularly left ventricular characteristics, indicates the model’s potential for medical image analysis applications. Conclusions: This work establishes diffusion models as a robust solution for cardiac MRI data augmentation, successfully generating anatomically accurate synthetic images that enhance downstream clinical applications while maintaining diagnostic fidelity. Full article

Object detection in aerial images can provide great support in traffic planning, national defense reconnaissance, hydrographic surveys, infrastructure construction, and other fields. Objects in aerial images are characterized by small pixel–area ratios, dense arrangements between objects, and arbitrary inclination angles. In response to these characteristics and problems, we improved the feature extraction network Inception-ResNet using the Fast Architecture Search (FAS) module and proposed a one-stage anchor-free rotation object detector. The structure of the object detector is simple and only consists of convolution layers, which reduces the number of model parameters. At the same time, the label sampling strategy in the training process is optimized to resolve the problem of insufficient sampling. Finally, a decoupled object detection head is used to separate the bounding box regression task from the object classification task. The experimental results show that the proposed method achieves mean average precision (mAP) of 82.6%, 79.5%, and 89.1% on the DOTA1.0, DOTA1.5, and HRSC2016 datasets, respectively, and the detection speed reaches 24.4 FPS, which can meet the needs of real-time detection. Full article

Deep learning models rely heavily on extensive training data, but obtaining sufficient real-world data remains a major challenge in clinical fields. To address this, we explore methods for generating realistic synthetic multivariate fall data to supplement limited real-world samples collected from three fall-related datasets: SmartFallMM, UniMib, and K-Fall. We apply three conventional time-series augmentation techniques, a Diffusion-based generative AI method, and a novel approach that extracts fall segments from public video footage of older adults. A key innovation of our work is the exploration of two distinct approaches: video-based pose estimation to extract fall segments from public footage, and Diffusion models to generate synthetic fall signals. Both methods independently enable the creation of highly realistic and diverse synthetic data tailored to specific sensor placements. To our knowledge, these approaches and especially their application in fall detection represent rarely explored directions in this research area. To assess the quality of the synthetic data, we use quantitative metrics, including the Fréchet Inception Distance (FID), Discriminative Score, Predictive Score, Jensen–Shannon Divergence (JSD), and Kolmogorov–Smirnov (KS) test, and visually inspect temporal patterns for structural realism. We observe that Diffusion-based synthesis produces the most realistic and distributionally aligned fall data. To further evaluate the impact of synthetic data, we train a long short-term memory (LSTM) model offline and test it in real time using the SmartFall App. Incorporating Diffusion-based synthetic data improves the offline F1-score by 7–10% and boosts real-time fall detection performance by 24%, confirming its value in enhancing model robustness and applicability in real-world settings. Full article

Background: Leigh’s Disease is a rare mitochondrial disorder primarily affecting the central nervous system, with frequent secondary cardiac manifestations such as hypertrophic and dilated cardiomyopathies. Early detection of cardiac complications is crucial for patient management, but manual interpretation of cardiac MRI is labour-intensive and subject to inter-observer variability. Methodology: We propose an integrated deep learning framework using cardiac MRI to automate the detection of cardiac abnormalities associated with Leigh’s Disease. Four CNN architectures—Inceptionv3, a custom 3-layer CNN, DenseNet169, and EfficientNetB2—were trained on preprocessed MRI data (224 × 224 pixels), including left ventricular segmentation, contrast enhancement, and gamma correction. Morphological features (area, aspect ratio, and extent) were also extracted to aid interpretability. Results: EfficientNetB2 achieved the highest test accuracy (99.2%) and generalization performance, followed by DenseNet169 (98.4%), 3-layer CNN (95.6%), and InceptionV3 (94.2%). Statistical morphological analysis revealed significant differences in cardiac structure between Leigh’s and non-Leigh’s cases, particularly in area (212,097 vs. 2247 pixels) and extent (0.995 vs. 0.183). The framework was validated using ROC (AUC = 1.00), Brier Score (0.000), and cross-validation (mean sensitivity = 1.000, std = 0.000). Feature embedding visualisation using PCA, t-SNE, and UMAP confirmed class separability. Grad-CAM heatmaps localised relevant myocardial regions, supporting model interpretability. Conclusions: Our deep learning-based framework demonstrated high diagnostic accuracy and interpretability in detecting Leigh’s disease-related cardiac complications. Integrating morphological analysis and explainable AI provides a robust and scalable tool for early-stage detection and clinical decision support in rare diseases. Full article

Automotive intelligent agents are increasingly collecting facial data for applications such as driver behavior monitoring and identity verification. These excessive collections of facial data bring serious risks of sensitive information leakage to autonomous driving. Facial information has been explicitly required to be anonymized, but the availability of most desensitized facial data is poor, which will greatly affect its application in autonomous driving. This paper proposes an automotive sensitive information anonymization method with high-quality generated facial images by considering the data availability under privacy protection. By comparing K-Same and Generative Adversarial Networks (GANs), this paper proposes a hierarchical self-attention mechanism in StyleGAN3 to enhance the feature perception of face images. The synchronous regularization of sample data is applied to optimize the loss function of the discriminator of StyleGAN3, thereby improving the convergence stability of the model. The experimental results demonstrate that the proposed facial desensitization model reduces the Frechet inception distance (FID) and structural similarity index measure (SSIM) by 95.8% and 24.3%, respectively. The image quality and privacy desensitization of the facial data generated by the StyleGAN3 model have been fully verified in this work. This research provides an efficient and robust facial privacy protection solution for autonomous driving, which is conducive to promoting the security guarantee of automotive data. Full article