Search Results (15,052)

In Sicily (Italy), respiratory syncytial virus (RSV), rhinovirus (HRV), and influenza virus triggered epidemics among children, resulting in an increase in acute respiratory tract infections (ARTIs). Our objective was to capture the epidemiology of respiratory infections in children, determining which pathogens were associated with respiratory infections following the lockdown and whether there were changes in the epidemiological landscape during the post-SARS-CoV-2 pandemic era. Materials and Methods: We analyzed multiplex respiratory viral PCR data (BioFire^® FilmArray^® Respiratory Panel 2.1 Plus) from 204 children presenting with respiratory symptoms and/or fever to our Unit of Pediatrics and Pediatric Emergency. Results: Viruses were predominantly responsible for ARTIs (99%), with RSV emerging as the most common agent involved in respiratory infections, followed by human rhinovirus/enterovirus and influenza A. RSV and rhinovirus were also the primary agents in coinfections. RSV predominated during winter months, while HRV/EV exhibited greater prevalence than RSV during the fall. Some viruses spread exclusively in coinfections (human coronavirus NL63, adenovirus, metapneumovirus, and parainfluenza viruses 1–3), while others primarily caused mono-infections (influenza A and B). SARS-CoV-2 was detected equally in both mono-infections (41%) and coinfections (59%). Conclusions: Our analysis underlines the predominance of RSV and the importance of implementing preventive strategies for RSV. Full article

The variety of sources of brachial plexus injuries (BPIs) and the severity and similarity of their clinical symptoms with those of other injuries make their differential diagnosis difficult. Enriching their diagnosis with objective high-sensitivity diagnostics such as clinical neurophysiology may lead to satisfactory treatment results, and magnetic stimulation (MEP) might be an advantageous addition to the diagnostic standard of electrical stimulation used in electroneurography (ENG). The asymptomatic side in BPI cases sometimes shows only subclinical neurological deficits; this study aimed to clarify the validity and utility of using MEP vs. ENG to detect neural conduction abnormalities. Twenty patients with a BPI and twenty healthy volunteers with matching demographic and anthropometric characteristics were stimulated at their Erb’s point in order to record the potentials evoked using magnetic and electrical stimuli to evaluate their peripheral motor neural transmission in their axillar, musculocutaneous, radial, and ulnar nerves. MEP was also used to verify the neural transmission in participants’ cervical roots following transvertebral stimulations, checking the compatibility and repeatability of the evoked potential recordings. The clinical assessment resulted in an average muscle strength of 3–1 (with a mean of 2.2), analgesia that mainly manifested in the C5–C7 spinal dermatomes, and a pain evaluation of 6–4 (mean of 5.4) on the symptomatic side using the Visual Analog Scale, with no pathological symptoms on the contralateral side. A comparison of the recorded potentials evoked with magnetic versus electrical stimuli revealed that the MEP amplitudes were usually higher, at p = 0.04–0.03, in most of the healthy volunteers’ recorded muscles than in those of the group of BPI patients, whose recordings showed that their CMAP and MEP amplitude values were lower on their more symptomatic than asymptomatic sides, at p = 0.04–0.009. In recordings following musculocutaneous and radial nerve electrical stimulation and ulnar nerve magnetic stimulation at Erb’s point, the values of the latencies were also longer on the patient’s asymptomatic side compared to those in the control group. The above outcomes prove the mixed axonal and demyelination natures of brachial plexus injuries. They indicate that different types of traumatic BPIs also involve the clinically asymptomatic side. Cases with predominantly median nerve lesions were detected in sensory nerve conduction studies (SNCSs). In 16 patients, electromyography revealed neurogenic damage to the deltoid and biceps muscles, with an active denervation process at work. The predominance of C5 and C6 brachial plexus injuries in the cervical root and upper/middle trunk of patients with BPI has been confirmed. A probable explanation for the bilateral symptoms of dysfunction detected via clinical neurophysiology methods in the examined BPI patients, who showed primarily unilateral damage, maybe the reaction of their internal neural spinal center’s organization. Even when subclinical, this may explain the poor BPI treatment outcomes that sometimes occur following long-term physical therapy or surgical treatment. Full article

Vehicle–infrastructure cooperative perception is becoming increasingly crucial for autonomous driving systems and involves leveraging infrastructure’s broader spatial perspective and computational resources. This paper introduces CoFormerNet, which is a novel framework for improving cooperative perception. CoFormerNet employs a consistent structure for both vehicle and infrastructure branches, integrating the temporal aggregation module and spatial-modulated cross-attention to fuse intermediate features at two distinct stages. This design effectively handles communication delays and spatial misalignment. Experimental results using the DAIR-V2X and V2XSet datasets demonstrated that CoFormerNet significantly outperformed the existing methods, achieving state-of-the-art performance in 3D object detection. Full article

In remote sensing image processing, the segmentation of clouds and their shadows is a fundamental and vital task. For cloud images, traditional deep learning methods often have weak generalization capabilities and are prone to interference from ground objects and noise, which not only results in poor boundary segmentation but also causes false and missed detections of small targets. To address these issues, we proposed a multi-branch attention fusion network (MAFNet). In the encoder section, the dual branches of ResNet50 and the Swin transformer extract features together. A multi-branch attention fusion module (MAFM) uses positional encoding to add position information. Additionally, multi-branch aggregation attention (MAA) in the MAFM fully fuses the same level of deep features extracted by ResNet50 and the Swin transformer, which enhances the boundary segmentation ability and small target detection capability. To address the challenge of detecting small cloud and shadow targets, an information deep aggregation module (IDAM) was introduced to perform multi-scale deep feature aggregation, which supplements high semantic information, improving small target detection. For the problem of rough segmentation boundaries, a recovery guided module (RGM) was designed in the decoder section, which enables the model to effectively allocate attention to complex boundary information, enhancing the network’s focus on boundary information. Experimental results on the Cloud and Cloud Shadow dataset, HRC-WHU dataset, and SPARCS dataset indicate that MAFNet surpasses existing advanced semantic segmentation techniques. Full article

In complex environments a single visible image is not good enough to perceive the environment, this paper proposes a novel dual-stream real-time detector designed for target detection in extreme environments such as nighttime and fog, which is able to efficiently utilise both visible and infrared images to achieve Fast All-Weatherenvironment sensing (FAWDet). Firstly, in order to allow the network to process information from different modalities simultaneously, this paper expands the state-of-the-art end-to-end detector YOLOv8, the backbone is expanded in parallel as a dual stream. Then, for purpose of avoid information loss in the process of network deepening, a cross-modal feature enhancement module is designed in this study, which enhances each modal feature by cross-modal attention mechanisms, thus effectively avoiding information loss and improving the detection capability of small targets. In addition, for the significant differences between modal features, this paper proposes a three-stage fusion strategy to optimise the feature integration through the fusion of spatial, channel and overall dimensions. It is worth mentioning that the cross-modal feature fusion module adopts an end-to-end training approach. Extensive experiments on two datasets validate that the proposed method achieves state-of-the-art performance in detecting small targets. The cross-modal real-time detector in this study not only demonstrates excellent stability and robust detection performance, but also provides a new solution for target detection techniques in extreme environments. Full article

Objective: The concentration of the placental circulating factor in early pregnancy is often extremely low, and the traditional prediction method cannot meet the clinical demand for early detection preeclampsia in high-risk gravida. It is of prime importance to seek an ultra-sensitive early prediction method. Methods: In this study, finite-different time-domain (FDTD) and Discrete Dipole Approximation (DDA) simulation, and electron beam lithography (EBL) methods were used to develop a bowtie nanoantenna (BNA) with the best field enhancement and maximum coupling efficiency. Bio-modification of the placental circulating factor (sFlt-1, PLGF) to the noble nanoparticles based on the amino coupling method were explored. A BNA LSPR biosensor which can specifically identify the placental circulating factor in preeclampsia was constructed. Results: The BNA LSPR biosensor can detect serum placental circulating factors without toxic labeling. Serum sFlt-1 extinction signal (Δλmax) in the preeclampsia group was higher than that in the normal pregnancy group (14.37 ± 2.56 nm vs. 4.21 ± 1.36 nm), p = 0.008, while the serum PLGF extinction signal in the preeclampsia group was lower than that in the normal pregnancy group (5.36 ± 3.15 nm vs. 11.47 ± 4.92 nm), p = 0.013. The LSPR biosensor detection results were linearly consistent with the ELISA kit. Conclusions: LSPR biosensor based on BNA can identify the serum placental circulating factor of preeclampsia with high sensitivity, without toxic labeling and with simple operation, and it is expected to be an early detection method for preeclampsia. Full article

The environment’s psychological impacts on humans have been long studied, but many questions remain unanswered. We conducted a micro-scale study to examine the relationships among the objective characteristics of urban environmental settings, people’s subjective perception of such settings, and the related psychological responses. We employed a geo-enabled survey tool to gather data on individuals’ perceptions of the immediate environment within their daily activity space. The psychological processes assessed included emotional and affective states such as perceived stress and happiness. The data points were mapped on a high-resolution aerial image, which was classified to derive quantitative properties to examine the dose-response relationship between environmental exposure and psychological responses. Our results showed negative correlations between the momentary stress level and the amount of environmental elements such as water, trees, and grass. Positive correlations were detected between stress level and the amount of parking lot and barren land, as well as the distance to buildings. In terms of perceived happiness, positive environmental factors included water, trees, and artificial surfaces, with all other elements having negative correlations. Most of the correlations examined were not strong correlations. This could be due to the significant differences in how individuals respond to environmental stimuli. Full article

Introduction: Periodontal disease is an infectious syndrome presenting inflammatory aspects. Radiographic evaluation is an essential complement to clinical assessment but has limitations such as the impossibility of assessing tissue inflammation. It seems essential to consider new exploration methods in clinical practice. Ultrasound of periodontal tissues could make it possible to visualize periodontal structures and detect periodontal diseases (periodontal pocket measurement and the presence of intra-tissue inflammation). Clinical Innovation Report: An ultrasound probe has been specially developed to explore periodontal tissues. The objective of this clinical innovation report is to present this device and expose its potential. Discussion: Various immediate advantages favor using ultrasound: no pain, no bleeding, faster execution time, and an image recording that can be replayed without having to probe the patient again. Ultrasound measurements of pocket depth appear to be as reliable and reproducible as those obtained by manual probing, as do tissue thickness measurements and the detection of intra-tissue inflammation. Conclusions: Ultrasound seems to have a broad spectrum of indications. Given the major advances offered by ultrasound imaging as a complementary aid to diagnosis, additional studies are necessary to validate these elements and clarify the potential field of application of ultrasound imaging in dentistry. Full article

Super-resolution (SR) is a technique that restores image details based on existing information, enhancing the resolution of images to prevent quality degradation. Despite significant achievements in deep-learning-based SR models, their application in underwater sonar scenarios is limited due to the lack of underwater sonar datasets and the difficulty in recovering texture details. To address these challenges, we propose a multi-scale generative adversarial network (SIGAN) for super-resolution reconstruction of underwater sonar images. The generator is built on a residual dense network (RDN), which extracts rich local features through densely connected convolutional layers. Additionally, a Convolutional Block Attention Module (CBAM) is incorporated to capture detailed texture information by focusing on different scales and channels. The discriminator employs a multi-scale discriminative structure, enhancing the detail perception of both generated and high-resolution (HR) images. Considering the increased noise in super-resolved sonar images, our loss function emphasizes the PSNR metric and incorporates the L2 loss function to improve the quality of the output images. Meanwhile, we constructed a dataset for side-scan sonar experiments (DNASI-I). We compared our method with the current state-of-the-art super-resolution image reconstruction methods on the public dataset KLSG-II and our self-built dataset DNASI-I. The experimental results show that at a scale factor of 4, the average PSNR value of our method was 3.5 higher than that of other methods, and the accuracy of target detection using the super-resolution reconstructed images can be improved to 91.4%. Through subjective qualitative comparison and objective quantitative analysis, we demonstrated the effectiveness and superiority of the proposed SIGAN in the super-resolution reconstruction of side-scan sonar images. Full article

Object detection in unmanned aerial vehicle (UAV) scenes is a challenging task due to the varying scales and complexities of targets. To address this, we propose a novel object detection model, AFE-YOLOv8, which integrates three innovative modules: the Multi-scale Nonlinear Fusion Module (MNFM), the Adaptive Feature Enhancement Module (AFEM), and the Receptive Field Expansion Module (RFEM). The MNFM introduces nonlinear mapping by exploiting the property that deformable convolution can dynamically adjust the shape of the convolution kernel according to the shape of the target, and it effectively enhances the feature extraction capability of the backbone network by integrating multi-scale feature maps from different mapping branches. Meanwhile, the AFEM introduces an adaptive fusion factor, and through the fusion factor, it adaptively integrates the small-target features contained in the feature maps of different detection branches into the small-target detection branch, thus enhancing the expression of the small-target features contained in the feature maps of the small-target detection branch. Furthermore, the RFEM expands the receptive field of the feature maps of the large- and medium-scale target detection branches through stacked convolution, so as to make the model’s receptive field cover the whole target, and thereby learn more rich and comprehensive features of the target. The experimental results demonstrate the superior performance of the proposed model compared to the baseline in detecting objects of various scales. On the VisDrone dataset, the proposed model achieves a 4.5% enhancement in mean average precision (mAP) and a 5.45% improvement in average precision at an IOU threshold of 0.5 (AP50). Additionally, ablation experiments conducted on the challenging DOTA dataset showcase the model’s robustness and generalization capabilities. Full article

The rising demand to efficiently acquire live production data has added more significance to automated monitoring and reporting within the industrial manufacturing sector. Real-time parts screening requiring repetitive human intervention for data input may not be a feasible solution to meet the demands of modern industrial automation. The objective of this study is to automatically classify and report on manufactured metal sheet parts. The metal components are mechanically suspended on an enamel paint-coating conveyor line in a household appliance manufacturing plant. At any given instant, the parts may not be in the exact coordinates within the desired area of interest and the classes of objects vary based on changing production requirements. To mitigate these challenges, this study proposes the use of a trained Mask R-CNN model to detect the objects and their associated class. Images are acquired in real-time using a video camera located next to the enamel coating line which are subsequently processed using the object detection algorithm for automated entry into the plant management information system. The highest achieved average precision obtained from the model was 98.27% with an overall accuracy of 98.24% using the proposed framework. The results surpassed the acceptable standard for the average precision of 97.5% as set by the plant production quality engineers. Full article

Background: Transabdominal ultrasound is used to detect fecal impaction, but the rectum is difficult to visualize without bladder urine or with gastrointestinal gas. Objective: We developed a transgluteal cleft approach that is unaffected by these factors and sought to determine if our ultrasound method could detect and classify fecal matter in the lower rectum using this approach. Methods: We classified ultrasound images from hospitalized patients into four groups: Group 1 (bowed and rock-like echogenic areas), Group 2 (irregular and cotton candy-like hyperechoic areas), Group 3 (flat and mousse-like hyperechoic areas), and Group 4 (linear echogenic areas in the lumen). Stool characteristics were classified as hard, normal, and muddy/watery. Sensitivity and specificity were determined based on fecal impaction and stool classification accuracy. Results: We obtained 129 ultrasound images of 23 patients. The sensitivity and specificity for fecal retention in the rectum were both 100.0%. The recall rates were 71.8% for Group 1, 93.1% for Group 2, 100.0% for Group 3, and 100.0% for Group 4. The precision rates were 96.6% for Group 1, 71.1% for Group 2, 88.9% for Group 3, and 100.0% for Group 4. Our method was 89.9% accurate overall. Conclusion: Transgluteal cleft approach ultrasound scanning can detect and classify fecal properties with high accuracy. Full article

In the field of object detection, the adversarial attack method based on generative adversarial network efficiently generates adversarial examples, thereby significantly reducing time costs. However, this approach overlooks the imperceptibility of perturbations in adversarial examples, resulting in poor visual performance and insufficient invisibility of the generated adversarial examples. To further enhance the imperceptibility of perturbations in adversarial examples, a method utilizing median filtering is proposed to address these generated perturbations. Experimental evaluations were conducted on the Pascal VOC dataset. The results demonstrate that, compared to the original image, there is an increase of at least 17.2% in the structural similarity index (SSIM) for generated adversarial examples. Additionally, the peak signal-to-noise ratio (PSNR) increases by at least 27.5%, while learned perceptual image patch similarity (LPIPS) decreases by at least 84.6%. These findings indicate that the perturbations in generated adversarial examples are more difficult to detect, with significantly improved imperceptibility and closer resemblance to the original image without compromising their high aggressiveness. Full article

Tactile texture sensors are designed to evaluate the sensations felt when a human touches an object. Prior studies have demonstrated the necessity for these sensors to have compliant ridges on their surfaces that mimic human fingerprints. These features enable the simulation of contact phenomena, especially friction and vibration, between human fingertips and objects, enhancing the tactile sensation evaluation. However, the ridges on tactile sensors are susceptible to abrasion damage from repeated use. To date, the healing function of abraded ridges has not been proposed, and its effectiveness needs to be demonstrated. In this study, we investigated whether the signal detection capabilities of a sensor with abraded epidermal ridges could be restored by healing the ridges using polyvinyl chloride plastisol as the sensor material. We developed a prototype tactile sensor with an embedded strain gauge, which was used to repeatedly scan roughness specimens. After more than 1000 measurements, we observed significant deterioration in the sensor’s output signal level. The ridges were then reshaped using a mold with a heating function, allowing the sensor to partially regain its original signal levels. This method shows potential for extending the operational lifespan of tactile texture sensors with compliant ridges. Full article

Fire is a significant security threat that can lead to casualties, property damage, and environmental damage. Despite the availability of object-detection algorithms, challenges persist in detecting fires, smoke, and humans. These challenges include poor performance in detecting small fires and smoke, as well as a high computational cost, which limits deployments. In this paper, we propose an end-to-end object detector for fire, smoke, and human detection based on Deformable DETR (DEtection TRansformer) called FSH-DETR. To effectively process multi-scale fire and smoke features, we propose a novel Mixed Encoder, which integrates SSFI (Separate Single-scale Feature Interaction Module) and CCFM (CNN-based Cross-scale Feature Fusion Module) for multi-scale fire, smoke, and human feature fusion. Furthermore, we enhance the convergence speed of FSH-DETR by incorporating a bounding box loss function called PIoUv2 (Powerful Intersection of Union), which improves the precision of fire, smoke, and human detection. Extensive experiments on the public dataset demonstrate that the proposed method surpasses state-of-the-art methods in terms of the (mean Average Precision), with and reaching 66.7% and 84.2%, respectively. Full article