uuid
int64 0
6k
| title
stringlengths 8
285
| abstract
stringlengths 22
4.43k
|
|---|---|---|
3,100 | Competitive interaction between smoking and chronic obstructive pulmonary disease for explaining renal function reduction in hypertensive patients | Chronic kidney disease is a risk factor for cardiovascular events. Smoking and chronic obstructive pulmonary disease (COPD) are risk factors for renal impairment. The aim of this study was to test the combined effect of smoking and COPD on renal function decline in hypertensives. We enrolled 1728 hypertensives stratified by smoking status and presence/absence of COPD. To test the mutual effect modification by both smoking and COPD and e-GFR, we performed crude and adjusted linear regression analyses, these latter taking into account potential confounders. Smokers displayed significantly lower e-GFR values than non-smokers (90 ± 24 vs. 121 ± 35 ml/min/1.73 m2); this difference was confirmed when comparing e-GFR values between patients with/without COPD (81 ± 17 vs. 109 ± 32 ml/min/1.73 m2). Smoking and COPD were directly and significantly interrelated (Cramer's V coefficient = 0.200; P = < 0.001). At interaction analyses, smoking significantly modified the effect of COPD on e-GFR and COPD significantly modified the effect of smoking on e-GFR, indicating a competitive interaction between smoking and COPD in the appearance of renal damage. e-GFR was 35 ml/min/1.73 m2 lower in patients with COPD than in those without; this reduction was of higher magnitude than that found between COPD and COPD-free patients among smokers (19 ml/min/1.73 m2). Smoking and COPD competitively interact in the appearance of renal function decline. These results suggest to screen for kidney damageboth smokers and COPD patients, especially those with both conditions. |
3,101 | Moderate resolution LAI prediction using Sentinel-2 satellite data and indirect field measurements in Sikkim Himalaya | The leaf area index (LAI) has been traditionally used as a photosynthetic variable. LAI plays an essential role in forest cover monitoring and has been identified as one of the important climate variables. However, due to challenges in field sampling, complex topography, and availability of cloud-free optical satellite data, LAI assessment on larger scale is still unexplored in the Sikkim Himalayan area. We used two optical instruments, digital hemispherical photography (DHP) and LAI-2200C, to assess the LAI across four different forests following 20 × 20 m2 elementary sampling units (ESUs) in the Himalayan state of Sikkim, India. The use of Sentinel-2 derived vegetation indices (VIs) demonstrated a better correlation with the DHP based LAI estimates than using LAI-2200C. Further, the combination of both reflectance bands and VIs were integrated to predict the LAI maps using random forest model. The temperate evergreen forests demonstrated the highest LAI value, while the predicted maps exhibited LAI maxima of 3.4. The estimated vs predicted LAI for DHP and LAI-2200C based estimation demonstrated reasonably good (R2 = 0.63 and R2 = 0.68, respectively) agreement. Further, improvements on the LAI prediction can be attempted by minimizing errors from the inherent field protocols, optimizing the density of field measurements, and representing heterogeneity. The recent rise of frequent forest fires in Sikkim Himalaya prompts for better understanding of fuel load in terms of surface fuel or canopy fuel that can be linked to LAI. The high-resolution LAI map could serve as input to forest fuel bed characterization, especially in seasonal forests with significant variations in green leaves and litter, thereby offering inputs for forest management in changing climate. |
3,102 | A practical guideline for performing a comprehensive transthoracic echocardiogram in the congenital heart disease patient: consensus recommendations from the British Society of Echocardiography | Transthoracic echocardiography is an essential tool in the diagnosis, assessment, and management of paediatric and adult populations with suspected or confirmed congenital heart disease. Congenital echocardiography is highly operator-dependent, requiring advanced technical acquisition and interpretative skill levels. This document is designed to complement previous congenital echocardiography literature by providing detailed practical echocardiography imaging guidance on sequential segmental analysis, and is intended for implementation predominantly, but not exclusively, within adult congenital heart disease settings. It encompasses the recommended dataset to be performed and is structured in the preferred order for a complete anatomical and functional sequential segmental congenital echocardiogram. It is recommended that this level of study be performed at least once on all patients being assessed by a specialist congenital cardiology service. This document will be supplemented by a series of practical pathology specific congenital echocardiography guidelines. Collectively, these will provide structure and standardisation to image acquisition and reporting, to ensure that all important information is collected and interpreted appropriately. |
3,103 | A comparison of temporomandibular disorder headache in young adults in relation to perceived stress | Chronic pain in the head and face region has a predicted prevalence of 20% in Europe, and is more common in women than men. The etiology of temporomandibular disorder (TMD) is multifactorial, and high levels of psychological stress amplify its symptoms. We were interested in how headache reported in RDC/TMD is associated with stress. Individuals with an average age of 18 years were included in the study. They were all volunteer participants in a research project. Clinical evaluation of each subject was performed using the RDC/TMD dual-axis diagnostic system. All participants filled out the PSS-10 questionnaire. A total of 138 individuals participated in the study, of which 107 were female. Headache was reported by 83 participants (59.4%), with females presenting higher scores on the PSS-10; this was statistically significant. A comparison between PSS-10 questionnaire results and headache level shows insignificant differences. However, the higher the stress level in the participant, the higher the headache score. Females are more susceptible to perceived stress, which can have an effect on TMD. |
3,104 | Breast tumor classification in ultrasound images using texture analysis and super-resolution methods | Ultrasound images can be used to detect tumors that do not appear in the mammograms of dense breasts. Several computer-aided diagnosis (CAD) systems based on this type of images have been proposed to detect tumors and discriminate between benign and malignant ones. To characterize those lesions, many of the aforementioned systems rely on texture analysis methods. However, speckle noise and artifacts that appear in ultrasound images may degrade their performance. To tackle this problem, and contrary to the state-of-the-art methods that utilize a single image of the breast, this paper proposes the use of a super-resolution approach that exploits the complementary information provided by multiple images of the same target. The proposed CAD system consists of four stages: super-resolution computation, extraction of the region of interest, feature extraction and classification. We have evaluated the performance of five texture methods with the proposed CAD system: gray level co-occurrence matrix features, local binary patterns, phase congruency-based local binary pattern, histogram of oriented gradients and pattern lacunarity spectrum. We show that our super-resolution based approach improves the performance of the evaluated texture methods and thus outperforms the state of the art in benign/malignant tumor classification. |
3,105 | Thermal-Aware Design for Approximate DNN Accelerators | Recent breakthroughs in Neural Networks (NNs) have made DNN accelerators ubiquitous and led to an ever-increasing quest on adopting them from Cloud to edge computing. However, state-of-the-art DNN accelerators pack immense computational power in a relatively confined area, inducing significant on-chip power densities that lead to intolerable thermal bottlenecks. Existing state of the art focuses on using approximate multipliers only to trade-off efficiency with inference accuracy. In this work, we present a thermal-aware approximate DNN accelerator design in which we additionally trade-off approximation with temperature effects towards designing DNN accelerators that satisfy tight temperature constraints. Using commercial multi-physics tool flows for heat simulations, we demonstrate how our thermal-aware approximate design reduces the temperature from 139 degrees C, in an accurate circuit, down to 79 degrees C. This enables DNN accelerators to fulfill tight thermal constraints, while still maximizing the performance and reducing the energy by around 75% with a negligible accuracy loss of merely 0.44% on average for a wide range of NN models. Furthermore, using physics-based transistor aging models, we demonstrate how reductions in voltage and temperature obtained by our approximate design considerably improve the circuit's reliability. Our approximate design exhibits around 40% less aging-induced degradation compared to the baseline design. |
3,106 | Multiple idiopathic cervical root resorption: A systematic review | The current literature on multiple idiopathic cervical root resorption (MICRR), a rare and aggressive form of external root resorption, is limited to case reports and series. Therefore, we performed a systematic review of this condition. A comprehensive search of PubMed, Embase, Web of science, Cochrane Library, CNKI, and WANFANG was conducted using key terms relevant to MICRR, supplemented by a grey literature search. Risk of bias was assessed using Cochrane's and Joanna Briggs Institute's tools. A total of 36 studies with 47 cases were included. MICRR is more common among younger females and may be related to hormonal changes and denosumab use. Initially, the premolars are usually affected but all permanent teeth may eventually be involved. Cone-beam computed tomography is recommended for diagnosis and assessment of resorptive lesions. The management is focused on complete removal and restoration of the resorptive tissue to maintain the tooth's structural integrity. However, MICRR usually has a poor prognosis. Due to its invasive and aggressive behavior, MICRR requires greater attention. |
3,107 | Drone Detection and Pose Estimation Using Relational Graph Networks | With the upsurge in use of Unmanned Aerial Vehicles (UAVs), drone detection and pose estimation by using optical sensors becomes an important research subject in cooperative flight and low-altitude security. The existing technology only obtains the position of the target UAV based on object detection methods. To achieve better adaptability and enhanced cooperative performance, the attitude information of the target drone becomes a key message to understand its state and intention, e.g., the acceleration of quadrotors. At present, most of the object 6D pose estimation algorithms depend on accurate pose annotation or a 3D target model, which costs a lot of human resource and is difficult to apply to non-cooperative targets. To overcome these problems, a quadrotor 6D pose estimation algorithm was proposed in this paper. It was based on keypoints detection (only need keypoints annotation), relational graph network and perspective-n-point (PnP) algorithm, which achieves state-of-the-art performance both in simulation and real scenario. In addition, the inference ability of our relational graph network to the keypoints of four motors was also evaluated. The accuracy and speed were improved significantly compared with the state-of-the-art keypoints detection algorithm. |
3,108 | An Overview of the State of the Art in Aircraft Prognostic and Health Management Strategies | Aircraft are complex engineering systems composed of many interconnected subsystems with possible uncertainties in their structure. They often function for a long number of flight hours under varying or harsh environments. Hence, prognostic and health management (PHM) of critical subsystems or components within the overall system is crucial for maintaining the safety and reliability of the aircraft. This article reviews the state of the art in aircraft failure prognostic. The main definitions and concepts are presented and discussed. In addition, a selected important failure in the representative aircraft components is outlined, and various categories of prognostic strategies are reviewed. Finally, some recommendations and directions for the most promising research to address the PHM problem in aircraft are outlined. |
3,109 | Robust optical flow estimation via edge preserving filtering | It is known that optical flow estimation techniques suffer from the issues of ill-defined edges and boundaries of the moving objects. Traditional variational methods for optical flow estimation are not robust to handle these issues since the local filters in these methods do not hold the robustness near the edges. In this paper, we propose a non-local total variation NLTV-L-1 optical flow estimation method based on robust weighted guided filtering. Specifically, first, the robust weighted guided filtering objective function is proposed to preserve motion edges. The proposed objective function is based on the linear model which is computationally efficient and edge-preserving in complex natural scenarios. Second, the proposed weighted guided filtering objective function is incorporated into the non-local total variation NLTV-L-1 energy function. Finally, the novel NLTV-L-1 optical flow method is performed using the coarse-to-fine process. Additionally, we modify some state-of-the-art variational optical flow estimation methods by the robust weighted guided filtering objective function to verify the performance on Middlebury, MPI-Sintel, and Foggy Zurich sequences. Experimental results show that the proposed method can preserve edges and improve the accuracy of optical flow estimation compared with several state-of-the-art methods. |
3,110 | Clinical, biochemical, neuroimaging and molecular findings of X-linked Adrenoleukodystrophy patients in South China | X-linked adrenoleukodystrophy is a common X-linked recessive peroxisomal disorder caused by the mutations in the ABCD1 gene. In this study, we analyzed 19 male patients and 9 female carriers with X-linked adrenoleukodystrophy in South China. By sequencing the ABCD1 gene, 13 different mutations were identified, including 7 novel mutations, and 6 known mutations, and 1 reported polymorphism. Mutation c.1180delG was demonstrated to be de novo mutation. 26.3 % (5/19) patients carried the deletion c.1415_16delAG, which may be the mutational hot spot in South China population. In addition, 73.7 % (14/19) patients were type of childhood cerebral adrenoleukodystrophy, 26.3 %(5/19) were in Addison only. Half of the childhood cerebral adrenoleukodystrophy patients had the adrenocortical insufficiency preceded the onset of neurological symptoms. Furthermore, 5 of 19 cases underwent hematopoietic stem cell transplantation. Our data showed that hematopoietic stem cell transplantation performed at an advanced stage of the cerebral X- linked adrenoleukodystrophy would accelerate the progression of the disease. Good clinical outcome achieved when hematopoietic stem cell transplantation performed at the very early stage of the disease. |
3,111 | Finite element analysis of electrical machines and transformers State of the art and future trends | Purpose - The purpose of this paper is to discuss the state of the art of finite element analysis of electrical machines and transformers. Electrical machines and transformers are prime examples of multi-physical systems involving electromagnetics, thermal issues, fluid dynamics, structural mechanics as well as acoustic phenomena. An accurate operational performance with different electrical and mechanical load situations is more and more evaluated using various numerical analysis methods including the couplings between the various physical domains. Therefore, numerical analysis methods are increasingly utilized not only for the verification of contractual values of existing machines, but also for the initial design process and for the design optimization of new machines. Design/methodology/approach - The finite element method is the most powerful numerical analysis method for such multi-physical devices. Since optimizations with respect to the overall performance and also the total manufacturing costs will become more important, the utilization of coupled multi-physical analyses is of growing interest. For the fast and powerful application of this numerical analysis method, special attention should be given to the requirements of these electromagnetic devices. Findings - Various methods of coupling the different physical domains of multi-field finite element analyses are described. Thereby, weakly coupled cascade algorithms can be used with most problems in the field of electrical machines and transformers. On the other hand, a prime objective is to derive comprehensive, multi-physical simulation models which are easily incorporated into design tools used by engineering professionals. Research limitations/implications - The development of robust and reliable computer-aided tools for an optimal design of multi-physical devices such electrical machines and transformers has to argue about the best possible coupling of various simulation methods. Special consideration shall be paid more and more to a treatment of uncertainties and tolerances by means of statistical and probabilistic approaches. Originality/value - The paper discusses state of the art of finite element analyses of the mentioned devices. Various optimized methods of modelling and analysis concerning the repetitive structure of electrical machines for electromagnetic analyses are compared with their advantages and drawbacks. Further, various methods of coupling the different domains of multi-field analyses in case of electrical machines and transformers are described. |
3,112 | Ancestry and BMI Influences on Facial Soft Tissue Depths for A Cohort of Chinese and Caucasoid Women in Dunedin, New Zealand | This study measured and assessed facial soft tissue depths (FSTDs) in adult female Chinese and New Zealand (NZ) Europeans (Caucasoids). Ultrasound was used to obtain depths at nine landmarks on 108 healthy subjects (51 Chinese, 57 NZ European), erect positioned, of same age group (18-29 years). Height and weight were also recorded. Statistical analysis focused on comparison of tissue depth between the two ancestry groups and the influence of Body Mass Index (BMI) (kg/m2). Results showed mean depth differences at Supra M2 and Infra M2 landmarks significantly greater for Chinese than Caucasoid women for all three BMI Classes (BMI<20, 20≤BMI<25, 25≤BMI<30), even BMI<20. For both groups BMI positively correlated with FSTD values at all landmarks except Labrale superius. This study enabled ancestry and BMI influence on FSTDs to be observed and compared for two distinct groups. Results add to knowledge about facial tissue depth variation. |
3,113 | Pattern recognition methodologies for pollen grain image classification: a survey | In a large number of scientific areas, such as immunology, forensics, paleoecology, and archeology, the study of pollen, i.e., palynology, plays an important role: from tracking climate changes, studying allergies, to forensic investigations or honey origin analysis. Since the mid-nineties of the last century, the idea for an automated solution to the problem of pollen identification and classification was formulated and since then, several attempts and proposals have been made and presented, based on different technologies, in particular in the field of Computer Vision. However, as of 2021 microscopic analyses are performed mainly manually by highly trained specialists, although the capabilities of artificial intelligence, especially Deep Neural Networks, are steadily increasing. In this work, we analyzed various state-of-the-art research work concerning pollen detection and classification and compared their methods and results. The problems, such as data accessibility, different methods of Machine Learning, and the intended applicability of the proposed solutions are explored. We also identified crucial issues that require further work and research. Our work will provide a thorough view on the current state of the art, its issues, and possibilities for the future. |
3,114 | Multi-label emotion classification in texts using transfer learning | Social media is a widespread platform that provides a massive amount of user-generated content that can be mined to reveal the emotions of social media users. This has many potential benefits, such as getting a sense of people's pulse on various events or news. Emotion classification from social media posts is challenging, especially when it comes to detecting multiple emotions from a short piece of text, as in multi-label classification problem. Most of the previous work on emotion detection has focused on deep neural networks such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) such as Long Short -Term Memory (LSTM) Networks. However, none of them has utilized multiple attention mechanisms and Recurrent Neural Networks (i.e., specialized attention networks for each emotion) nor utilized the recently introduced Transformer Networks such as XLNet, DistilBERT, and RoBERTa for the task of classifying emotions with multiple labels. The proposed multiple attention mechanism reveals the contribution of each word on each emotion, which has not been investigated before. In this study, we investigate both the use of LSTMs and the fine-tuning of Transformer Networks through Transfer Learning along with a single-attention network and a multiple-attention network for multi-label emotion classification. The experimental results show that our novel transfer learning models using pre-trained transformers with and without multiple attention mechanisms were able to outperform the current state-of-the-art accuracy (58.8% -Baziotis et al., 2018) in the SemEval-2018 Task-1C dataset. Our best-performing RoBERTa-MA (RoBERTa-Multi-attention) model outperformed the state-of-the-art and achieved 62.4% accuracy (3.6% gain over the state-of-the-art) on the challenging SemEval-2018 E-c: Detecting Emotions (multi-label classification) dataset for English. Moreover, the XLNet-MA (XLNet-Multi-attention) model outperformed the other proposed models by achieving 45.6% accuracy on the Ren-CECps dataset for Chinese. |
3,115 | Comparative Validation of Single-Shot Optical Techniques for Laparoscopic 3-D Surface Reconstruction | Intra-operative imaging techniques for obtaining the shape and morphology of soft-tissue surfaces in vivo are a key enabling technology for advanced surgical systems. Different optical techniques for 3-D surface reconstruction in laparoscopy have been proposed, however, so far no quantitative and comparative validation has been performed. Furthermore, robustness of the methods to clinically important factors like smoke or bleeding has not yet been assessed. To address these issues, we have formed a joint international initiative with the aim of validating different state-of-the-art passive and active reconstruction methods in a comparative manner. In this comprehensive in vitro study, we investigated reconstruction accuracy using different organs with various shape and texture and also tested reconstruction robustness with respect to a number of factors like the pose of the endoscope as well as the amount of blood or smoke present in the scene. The study suggests complementary advantages of the different techniques with respect to accuracy, robustness, point density, hardware complexity and computation time. While reconstruction accuracy under ideal conditions was generally high, robustness is a remaining issue to be addressed. Future work should include sensor fusion and in vivo validation studies in a specific clinical context. To trigger further research in surface reconstruction, stereoscopic data of the study will be made publically available at www.open-CAS.com upon publication of the paper. |
3,116 | A spatial-frequency-temporal 3D convolutional neural network for motor imagery EEG signal classification | Motor imagery (MI) EEG signal classification is a critical issue for brain-computer interface (BCI) systems. In traditional MI EEG machine learning algorithms, feature extraction and classification often have different objective functions, thus resulting in information loss. To solve this problem, a novel spatial-frequency-temporal (SFT) 3D CNN model is proposed. Specifically, the energies of EEG signals located in multiple local SFT ranges are extracted to obtain a novel 3D MI EEG feature representation, and a novel 3D CNN model is designed to simultaneously learn the complex MI EEG features in the entire SFT domains and carry out classification. An extensive experimental study is implemented on two public EEG datasets to evaluate the effectiveness of our method. For BCI Competition III Dataset IVa, the average accuracy rate of five subjects obtained by the proposed method reaches 86.6% and yields 4.1% improvement over the state-of-the-art filter band common spatial pattern (FBCSP) method. For BCI Competition III dataset IIIa, by achieving an average accuracy rate of 91.85%, the proposed method outperforms the state-of-the-art dictionary pair learning (DPL) method by 4.44%. |
3,117 | Particulate matter exposure exacerbates cellular damage by increasing stress granule formation in respiratory syncytial virus-infected human lung organoids | Exposure to atmospheric particulate matter (PM) increases morbidity and mortality in respiratory diseases by causing various adverse health effects; however, the effects of PM exposure on cellular stress under virus-infected conditions remain unclear. The effects of PM under 10 μm (PM10) and diesel PM (DPM) on respiratory syncytial virus (RSV) infection were investigated in human two-dimensional lung epithelial cells and human three-dimensional lung organoids mimicking the lung tissue. We evaluated the formation of stress granules, which are important in cellular adaptation to various stress conditions. Furthermore, we investigated the effects of repeated exposure to PM10 and DPM on DNA damage and cell death during viral infection. PM10 and DPM did not cause stress granule formation in the absence of RSV infection but drastically increased stress granule formation and signal transduction during RSV infection in human lung epithelial cells and human lung organoids. Further, repeated exposure to PM10 and DPM caused cell death by severely damaging DNA under RSV infection conditions. Thus, PM10 and DPM induce severe lung toxicity under stress conditions, such as viral infection, suggesting that the effects of PMs under various stressful conditions should be examined to accurately predict the lung toxicity of PM. |
3,118 | Biomimetic Patch with Wicking-Breathable and Multi-mechanism Adhesion for Bioelectrical Signal Monitoring | Wearable bioelectrical monitoring devices can provide long-term human health information such as electrocardiogram and other physiological signals. It is a crucial part of the remote medical system. These can provide prediction for the diagnosis and treatment of cardiovascular disease and access to timely treatment. However, the patch comfort of the wearable monitoring devices in long-term contact with the skin have been a technical bottleneck of the hardware. In this study, the biomimetic patch with wicking-breathable and multi-mechanism adhesion performance to achieve adaptability and comfortability to human skin has been reported. The patch was designed based on a conical through-hole and hexagonal microgroove to directionally transport sweat from skin to air which gives the patch the breathable performance. The breathable and drainage capability of the biomimetic patch was experimentally verified by analyzing the conical through-hole and hexagonal microgroove with the structural mechanism of wicking. Multi-mechanism adhesion of the Ag/Ni microneedle array and PDMS-t adhesion material ensures the stability of patch signal acquisition. This study provides a new way for enhancing the breathability and adaptability of the patch to realize accurate bioelectrical signal monitoring under sweat conditions on human skin. |
3,119 | Barriers and bridges on water management in rural Mexico: from water-quality monitoring to water management at the community level | Access to sufficient water of suitable quality represents a challenge for achieving several dimensions of sustainable development. Currently, water access is restricted to three of 10 persons globally. In rural areas of Mexico and other low-income countries, coverage could be even less due to the absence of formal supply; thus, rural communities usually perform water management. Surrounding community-based water management, various socio-ecological interactions emerge that determine access to water. Access to water will depend on the obstacles or capacities that arise within the socio-ecological system in which the community is immersed. This work identifies barriers and bridges to water access in a rural environment through mixed methods. The article draws on three case studies in southeastern Mexico by analyzing 90 questionnaires conducted at the household level and three focus groups in parallel with water quality analysis and its relationship with management practices. The barriers and bridges were classified into six water access challenges: (i) access to water in a sufficient quantity, (ii) access to water of adequate quality, (iii) access to water for household crop irrigation, (iv) hygiene and sanitation facilities, (v) collective organization, and (vi) climate variability. The main findings indicate that households' water quantity and quality show deficiencies due to the lack of formal infrastructure and represent a health risk. Water fetching has the highest impact on women and children in poor rural areas, and it is a significant barrier to sustainable development. In contrast, the collective organization proved to be an essential bridge for water access in these communities. |
3,120 | Synthetic-to-Real Domain Adaptation Joint Spatial Feature Transform for Stereo Matching | Most deep learning-based state-of-the-art stereo matching methods significantly depend on large-scale datasets. However, it is implausible to collect sufficient real-world samples with dense and clear ground-truth disparity maps in practice. Although synthetic datasets' appearance has alleviated the demand for extensive real data, there is a domain shift between synthetic and real sets. To tackle this problem, we propose an individually trained synthetic-to-real domain adaptation (SDA) network that maps synthetic images into the real domain. Specifically, our approach translates the data style from synthetic domain to real domain while maintaining the content and the spatial information. First, edge cues are leveraged to guide domain adaptation in preserving the spatial consistency between input and the generated image. Second, we combine the spatial feature transform (SFT) layer to effectively fuse features from the edge map and the source image. Extensive experiments demonstrate that: 1) when only trained on synthetic data and generalized to real data, our model evidently outperforms many state-of-the-art domain adaptation methods; 2) our translated synthetic datasets (TSD) help to improve the generalization capability of any stereo matching CNNs. Codes and data will be available at https://github.com/Archaic-Atom/SDA_network. |
3,121 | Optimal Discriminative Projection for Sparse Representation-Based Classification via Bilevel Optimization | Recently, sparse representation-based classification (SRC) has been widely studied and has produced state-of-the-art results in various classification tasks. Learning useful and computationally convenient representations from complex redundant and highly variable visual data is crucial for the success of SRC. However, how to find the best feature representation to work with SRC remains an open question. In this paper, we present a novel discriminative projection learning approach with the objective of seeking a projection matrix such that the learned low-dimensional representation can fit SRC well and that it has well discriminant ability. More specifically, we formulate the learning algorithm as a bilevel optimization problem, where the optimization includes an l(1)-norm minimization problem in its constraints. Through the bilevel optimization model, the relationship between sparse representation and the desired feature projection can be explicitly exploited during the learning process. Therefore, SRC can achieve a better performance in the transformed subspace. The optimization model can be solved by using a stochastic gradient ascent algorithm, and the desired gradient is computed using implicit differentiation. Furthermore, our method can be easily extended to learn a dictionary. The extensive experimental results on a series of benchmark databases show that our method outperforms many state-of-the-art algorithms. |
3,122 | Randomized Error Removal for Online Spread Estimation in High-Speed Networks | Flow spread measurement provides fundamental statistics that can help network operators better understand flow characteristics and traffic patterns with applications in traffic engineering, cybersecurity and quality of service. Past decades have witnessed tremendous performance improvement for single-flow spread estimation. However, when dealing with numerous flows in a packet stream, it remains a significant challenge to measure per-flow spread accurately while reducing memory footprint. The goal of this paper is to introduce new multi-flow spread estimation designs that incur much smaller processing overhead and query overhead than the state of the art, yet achieves significant accuracy improvement in spread estimation. We formally analyze the performance of these new designs. We implement them in both hardware and software, and use real-world data traces to evaluate their performance in comparison with the state of the art. The experimental results show that our best sketch significantly improves over the best existing work in terms of estimation accuracy, packet processing throughput, and online query throughput. |
3,123 | Maximum margin object tracking with weighted circulant feature maps | Support vector machine (SVM) based tracking algorithms training with dense circulant samples have shown favourable performance due to its strong discriminative power and high efficiency. However, the challenges caused by the circulant sampling remain unaddressed. In this study, the authors give each training sample a weight based on their accuracy to reduce the influence of inaccurate samples. Moreover, they reform the SVM model with weighted circulant training samples. Secondly, they advocate an efficient solution by using the property of circulant matrices to solve the learning problem. Thirdly, a model update strategy is introduced to prevent the tracking models polluted by wrong samples. Experimental results on large benchmark datasets with 50 and 100 video sequences demonstrate that the authors' tracking algorithms achieve state-of-art performance in terms of precision and accuracy. In addition, their tracker runs in real time. |
3,124 | Severe Vitamin B12 Deficiency Presenting as Pancytopenia, Hemolytic Anemia, and Paresthesia: Could Your B12 Be Any Lower? | Although severe vitamin B12 deficiency is rare in the United States, recent increases in the adoption of vegan lifestyles have led to a significant rise in the rates of B12 deficiency, along with its hematologic and neurologic sequelae, the latter of which is often irreversible. We describe a case of a 39-year-old male who presented with a several-month history of progressively worsening word-finding difficulties, shortness of breath, and a four-day history of bilateral hand numbness and tingling. Laboratory data revealed pancytopenia with profound anemia. Markers of hemolysis were positive, including elevated indirect bilirubin, disproportionately elevated lactate dehydrogenase (LDH), low haptoglobin, negative direct anticoagulant test, and hypoproliferative reticulocyte index. Blood smear revealed hypersegmented neutrophils and macrocytosis. Vitamin B12 levels were undetectable, and anti-intrinsic factor and parietal cell antibodies were negative. A thorough history revealed a 20-year history of strict veganism without B12 supplementation. He was transfused with packed red blood cells and started on subcutaneous B12 injections with rapid improvement of his symptoms. Early recognition of B12 deficiency causing the constellation of pancytopenia, hemolytic anemia, and neurologic symptoms is vital in preventing irreversible neurologic sequelae. This case also highlights the importance of accurate history taking to aid in early diagnosis of B12 deficiency, especially in the context of rising rates of veganism in the United States. |
3,125 | The Layer-Wise Training Convolutional Neural Networks Using Local Loss for Sensor-Based Human Activity Recognition | Recently, deep learning, which are able to extract automatically features from data, has achieved state-of-the-art performance across a variety of sensor based human activity recognition (HAR) tasks. However, the existing deep neural networks are usually trained with a global loss, and all hidden layer weights have to be always kept in memory before the forward and backward pass has completed. The backward locking phenomenon prevents the reuse of memory, which is a crucial limitation for wearable activity recognition. In the paper, we proposed a layer-wise convolutional neural networks (CNN) with local loss for the use of HAR task. To our knowledge, this paper is the first that uses local loss based CNN for HAR in ubiquitous and wearable computing arena. We performed experiments on five public HAR datasets including UCI HAR dataset, OPPOTUNITY dataset, UniMib-SHAR dataset, PAMAP dataset, and WISDM dataset. The results show that local loss works better than global loss for tested baseline architectures. At no extra cost, the local loss can approach the state-of-the-arts on a variety of HAR datasets, even though the number of parameters was smaller. We believe that the layer-wise CNN with local loss can be used to update the existing deep HAR methods. |
3,126 | A systematic review of current status and challenges of vaccinating children against SARS-CoV-2 | The coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has inflicted immense damage to countries, economies and societies worldwide. Authorized COVID-19 vaccines based on different platforms have been widely inoculated in adults, showing up to 100% immunogenicity with significant efficacy in preventing SARS-CoV-2 infections and the occurrence of severe COVID-19. It has also greatly slowed the evolution of SARS-CoV-2 variants, as shown in clinical trials and real-world evidence. However, the total dosage of COVID-19 vaccines for children is much smaller than that for adults due to limitations from parental concern of vaccine safety, presenting a potential obstacle in ending the COVID-19 pandemic. SARS-CoV-2 not only increases the risk of severe multisystem inflammatory syndrome (MIS-C) in children, but also negatively affects children's psychology and academics, indirectly hindering the maintenance and progress of normal social order. Therefore, this article examines the clinical manifestations of children infected with SARS-CoV-2, the status of vaccination against COVID-19 in children, vaccination-related adverse events, and the unique immune mechanisms of children. In particular, the necessity and challenges of vaccinating children against SARS-CoV-2 were highlighted from the perspectives of society and family. In summary, parental hesitancy is unnecessary as adverse events after COVID-19 vaccination have been proven to be infrequent, comprise of mild symptoms, and have a good prognosis. |
3,127 | The political system through a partisan lens: Within-person changes in support for political parties precede political system attitudes | Although political party support and attitudes towards the political system are closely related, the temporal ordering of these associations is unclear. Indeed, prior research identifies both partisan-led change in system attitudes and system attitude-led change in party support. Using a ten-year (2010-2020) national probability sample of New Zealand adults (N = 66,359), we test these associations by modelling the within-person cross-lagged effects between conservative and liberal party support, and political system justification. During conservative-led governments, increases in conservative party support predicted increases in political system justification more strongly than vice versa. The 2017 shift to a liberal-led government was met with an immediate reversal of the effects of party support on system justification, but the effect of system justification on party support took a full year to reverse. These results demonstrate people's perceptions of the fairness of the political system depend on their support for the party in power. |
3,128 | Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning | Remarkable progress has been made in image recognition, primarily due to the availability of large-scale annotated datasets and deep convolutional neural networks (CNNs). CNNs enable learning data-driven, highly representative, hierarchical image features from sufficient training data. However, obtaining datasets as comprehensively annotated as ImageNet in the medical imaging domain remains a challenge. There are currently three major techniques that successfully employ CNNs to medical image classification: training the CNN from scratch, using off-the-shelf pre-trained CNN features, and conducting unsupervised CNN pre-training with supervised fine-tuning. Another effective method is transfer learning, i.e., fine-tuning CNN models pre-trained from natural image dataset to medical image tasks. In this paper, we exploit three important, but previously understudied factors of employing deep convolutional neural networks to computer-aided detection problems. We first explore and evaluate different CNN architectures. The studied models contain 5 thousand to 160 million parameters, and vary in numbers of layers. We then evaluate the influence of dataset scale and spatial image context on performance. Finally, we examine when and why transfer learning from pre-trained ImageNet (via fine-tuning) can be useful. We study two specific computer-aided detection (CADe) problems, namely thoraco-abdominal lymph node (LN) detection and interstitial lung disease (ILD) classification. We achieve the state-of-the-art performance on the mediastinal LN detection, and report the first five-fold cross-validation classification results on predicting axial CT slices with ILD categories. Our extensive empirical evaluation, CNN model analysis and valuable insights can be extended to the design of high performance CAD systems for other medical imaging tasks. |
3,129 | Platelet Reactivity: Is There a Role to Switch? | Antiplatelet agents are essential to prevent thrombotic events in patients with coronary artery disease, especially in those with acute coronary syndrome or undergoing percutaneous coronary intervention (PCI). However, the benefits of antiplatelet therapy always come at a price of increased risk for bleeding, and the clinical values of antiplatelet strategies depend on this characteristic benefit/risk ratio. Platelet function testing aiming at determining an individual's response to the administered agent was hoped to help balance bleeding and thrombosis in order to maximize benefit/risk ratio. However, randomized trials failed to demonstrate an improved clinical outcome of a platelet function-based treatment selection and consequently platelet function testing has not become a routine part of the management of antiplatelet therapies. This review aims to discuss results and shortcomings of available trials and registries regarding the potential role of platelet reactivity testing in guiding antiplatelet treatment selection in patients undergoing PCI. |
3,130 | Missing value imputation through shorter interval selection driven by Fuzzy C-Means clustering | The presence of missing data is a common and pivotal issue, which generally leads to a serious decrease of data quality and thus indicates the necessity to effectively handle missing data. In this paper, we propose a missing value imputation approach driven by Fuzzy C-Mean clustering to improve the classification accuracy by referring only to the known feature values of some selected instances. In particular, the missing values for each instance are imputed by selecting a shorter interval based on the cluster membership value within the certain threshold limit of each feature, while using a short interval is considered to improve the imputation effectiveness and get more accurate estimation of the values in comparison with using a long interval. Our method is evaluated through comparing with state-of-the-art imputation methods on UCI datasets. The experimental results demonstrate that the proposed approach performs closely to or better than those state-of-the-art imputation methods. |
3,131 | Enhanced Multi-Task Learning Architecture for Detecting Pedestrian at Far Distance | Existing pedestrian detection methods suffer from performance degradation in the presence of small-scale pedestrians who are positioned at far distance from the camera. We present a pedestrian detection framework that is not only robust to small- and large-scale pedestrians, but is also significantly faster than state-of-the-art methods. The proposed framework incorporates semantic segmentation to confidence modules for RPN (Region Proposal Network) head and R-FCN (Region-based Fully Convolutional Networks) head, and a cascaded R-FCN head. The semantic segmentation confidence is extracted and utilized as auxiliary classification prior knowledge for RPN proposal selection and R-FCN head prediction. Finally, the cascaded R-FCN head progressively refine the pedestrian prediction accuracy with negligible computation overhead. The proposed framework is also capable of maintaining high detection performance on down-sampled input images, which leads to further reduction in overall computational complexity. Experiment results on CityPersons and MOT17Det datasets show that the proposed framework achieves competitive detection performance with about 3x speedup over state-of-the-art methods. |
3,132 | Guided Soft Attention Network for Classification of Breast Cancer Histopathology Images | An attention guided convolutional neural network (CNN) for the classification of breast cancer histopathology images is proposed. Neural networks are generally applied as black box models and often the network's decisions are difficult to interpret. Making the decision process transparent, and hence reliable is important for a computer-assisted diagnosis (CAD) system. Moreover, it is crucial that the network's decision be based on histopathological features that are in agreement with a human expert. To this end, we propose to use additional region-level supervision for the classification of breast cancer histopathology images using CNN, where the regions of interest (RoI) are localized and used to guide the attention of the classification network simultaneously. The proposed supervised attention mechanism specifically activates neurons in diagnostically relevant regions while suppressing activations in irrelevant and noisy areas. The class activation maps generated by the proposed method correlate well with the expectations of an expert pathologist. Moreover, the proposed method surpasses the state-of-the-art on the BACH microscopy test dataset (part A) with a significant margin. |
3,133 | A review of magnetostatic moment method | This paper proposes a review of the magnetostatic moments method (MoM) applied to model electromagnetic devices. This method is now well-known for its "light weight" and its simplicity of implementation. Its main advantages are the nonrequirement of an air region mesh and a coarse mesh of the ferromagnetic material. It leads to very fast resolution and very accurate field, force, and moment computations. The paper proposes a state of the art of this approach and shows some efficient realizations. |
3,134 | Factors controlling sediment yield at the catchment scale in NW Mediterranean geoecosystems | Purpose This study aimed to (1) increase understanding of the relation between sediment yield and environmental variables at the catchment scale; (2) test and validate existing and newly developed regression equations for prediction of sediment yield; and (3) identify how better predictions may be obtained. Materials and methods A correlation and regression analysis was performed between sediment yield and over 40 environmental variables for 61 Spanish catchments. Variables were selected based on availability and expected relation with diverse soil erosion and sediment transport processes. For comparison, the Area Relief Temperature (ART) sediment delivery model was applied to the same catchments. Sediment yield estimates obtained from reservoir surveys were used for model calibration and validation. Results and discussion Catchment area, catchment perimeter, stream length, relief ratio, Modified Fournier Index, the RUSLE's R factor, and catchments percentage with poor vegetation cover showed highest correlations with sediment yield. Stepwise linear regression revealed that variables representing topography, climate, vegetation, lithology, and soil characteristics are required for the best prediction equation. Although calibration results were relatively good, validation showed that the models were unstable and not suitable for extrapolation to other catchments. Reasons for this unstable model performance include (1) lack of detail and quality of the data sources; (2) large variation in catchment characteristics; (3) insufficient representation of all relevant erosion and sediment transport processes; and (4) the presence of nonlinear relations between sediment yield and environmental variables. The nonlinear ART model performed relatively well but systematically overpredicted sediment yield. A model reflecting human impacts, including dams and conservation measures, is expected to provide better results. This, however, requires significantly more input data. Conclusions Although important insight is obtained into the relation between sediment yield and environmental factors, prediction of sediment yield at the catchment scale requires alternative approaches. More detailed information is required on land cover (change), and the effect of soil conservation measures. Validation of regression equations is a necessity, and better predictions are obtained by nonlinear models. |
3,135 | Delineating the transcriptional landscape and clonal diversity of virus-specific CD4+ T cells during chronic viral infection | Although recent evidence indicates that CD4+ T cells responding to chronic viral infection are functionally heterogenous, our understanding of the developmental relationships between these subsets, and a determination of how their transcriptional landscape compares to their acute infection counterparts remains unclear. Additionally, whether cell-intrinsic factors such as TCR usage influence CD4+ T cell fate commitment during persistent infection has not previously been studied. Herein, we perform single-cell RNA sequencing (scRNA-seq) combined with single-cell T cell receptor sequencing (scTCR-seq) on virus-specific CD4+ T cells isolated from mice infected with chronic lymphocytic choriomeningitis virus (LCMV) infection. We identify several transcriptionally distinct states among the Th1, Tfh, and memory-like T cell subsets that form at the peak of infection, including the presence of a previously unrecognized Slamf7+ subset with cytolytic features. We further show that the relative distribution of these populations differs substantially between acute and persistent LCMV infection. Moreover, while the progeny of most T cell clones displays membership within each of these transcriptionally unique populations, overall supporting a one cell-multiple fate model, a small fraction of clones display a biased cell fate decision, suggesting that TCR usage may impact CD4+ T cell development during chronic infection. Importantly, comparative analyses further reveal both subset-specific and core gene expression programs that are differentially regulated between CD4+ T cells responding to acute and chronic LCMV infection. Together, these data may serve as a useful framework and allow for a detailed interrogation into the clonal distribution and transcriptional circuits underlying CD4+ T cell differentiation during chronic viral infection. |
3,136 | Clinical practice guidelines for duodenal cancer 2021 | Duodenal cancer is considered to be a small intestinal carcinoma in terms of clinicopathology. In Japan, there are no established treatment guidelines based on sufficient scientific evidence; therefore, in daily clinical practice, treatment is based on the experience of individual physicians. However, with advances in diagnostic modalities, it is anticipated that opportunities for its detection will increase in future. We developed guidelines for duodenal cancer because this disease is considered to have a high medical need from both healthcare providers and patients for appropriate management. These guidelines were developed for use in actual clinical practice for patients suspected of having non-ampullary duodenal epithelial malignancy and for patients diagnosed with non-ampullary duodenal epithelial malignancy. In this study, a practice algorithm was developed in accordance with the Minds Practice Guideline Development Manual 2017, and Clinical Questions were set for each area of epidemiology and diagnosis, endoscopic treatment, surgical treatment, and chemotherapy. A draft recommendation was developed through a literature search and systematic review, followed by a vote on the recommendations. We made decisions based on actual clinical practice such that the level of evidence would not be the sole determinant of the recommendation. This guideline is the most standard guideline as of the time of preparation. It is important to decide how to handle each case in consultation with patients and their family, the treating physician, and other medical personnel, considering the actual situation at the facility (and the characteristics of the patient). |
3,137 | Treatment with an antigen-specific dual microparticle system reverses advanced multiple sclerosis in mice | Antigen-specific therapies hold promise for treating autoimmune diseases such as multiple sclerosis while avoiding the deleterious side effects of systemic immune suppression due to delivering the disease-specific antigen as part of the treatment. In this study, an antigen-specific dual-sized microparticle (dMP) treatment reversed hind limb paralysis when administered in mice with advanced experimental autoimmune encephalomyelitis (EAE). Treatment reduced central nervous system (CNS) immune cell infiltration, demyelination, and inflammatory cytokine levels. Mechanistic insights using single-cell RNA sequencing showed that treatment impacted the MHC II antigen presentation pathway in dendritic cells, macrophages, B cells, and microglia, not only in the draining lymph nodes but also strikingly in the spinal cord. CD74 and cathepsin S were among the common genes down-regulated in most antigen presenting cell (APC) clusters, with B cells also having numerous MHC II genes reduced. Efficacy of the treatment diminished when B cells were absent, suggesting their impact in this therapy, in concert with other immune populations. Activation and inflammation were reduced in both APCs and T cells. This promising antigen-specific therapeutic approach advantageously engaged essential components of both innate and adaptive autoimmune responses and capably reversed paralysis in advanced EAE without the use of a broad immunosuppressant. |
3,138 | Region-Enhancing Network for Semantic Segmentation of Remote-Sensing Imagery | Semantic segmentation for high-resolution remote-sensing imagery (HRRSI) has become increasingly popular in machine vision in recent years. Most of the state-of-the-art methods for semantic segmentation of HRRSI usually emphasize the strong learning ability of deep convolutional neural network to model the contextual relationship in the image, which takes too much consideration on every pixel in images and subsequently causes the problem of overlearning. Annotation errors and easily confused features can also lead to the confusion problem while using the pixel-based methods. Therefore, we propose a new semantic segmentation network-the region-enhancing network (RE-Net)-to emphasize the regional information instead of pixels to solve the above problems. RE-Net introduces the regional information into the base network, to enhance the regional integrity of images and thus reduce misclassification. Specifically, the regional context learning procedure (RCLP) can learn the context relationship from the perspective of regions. The region correcting procedure (RCP) uses the pixel aggregation feature to recalibrate the pixel features in each region. In addition, another simple intra-network multi-scale attention module is introduced to select features at different scales by the size of the region. A large number of comparative experiments on four different public datasets demonstrate that the proposed RE-Net performs better than most of the state-of-the-art ones. |
3,139 | Tensor Factorization for Low-Rank Tensor Completion | Recently, a tensor nuclear norm (TNN) based method was proposed to solve the tensor completion problem, which has achieved state-of-the-art performance on image and video inpainting tasks. However, it requires computing tensor singular value decomposition (t-SVD), which costs much computation and thus cannot efficiently handle tensor data, due to its natural large scale. Motivated by TNN, we propose a novel low-rank tensor factorization method for efficiently solving the 3-way tensor completion problem. Our method preserves the low-rank structure of a tensor by factorizing it into the product of two tensors of smaller sizes. In the optimization process, our method only needs to update two smaller tensors, which can be more efficiently conducted than computing t-SVD. Furthermore, we prove that the proposed alternating minimization algorithm can converge to a Karush-Kuhn-Tucker point. Experimental results on the synthetic data recovery, image and video inpainting tasks clearly demonstrate the superior performance and efficiency of our developed method over state-of-the-arts including the TNN and matricization methods. |
3,140 | Degraded Image Semantic Segmentation With Dense-Gram Networks | Degraded image semantic segmentation is of great importance in autonomous driving, highway navigation systems, and many other safety-related applications and it was not systematically studied before. In general, image degradations increase the difficulty of semantic segmentation, usually leading to decreased semantic segmentation accuracy. Therefore, performance on the underlying clean images can be treated as an upper bound of degraded image semantic segmentation. While the use of supervised deep learning has substantially improved the state of the art of semantic image segmentation, the gap between the feature distribution learned using the clean images and the feature distribution learned using the degraded images poses a major obstacle in improving the degraded image semantic segmentation performance. The conventional strategies for reducing the gap include: 1) Adding image-restoration based preprocessing modules; 2) Using both clean and the degraded images for training; 3) Fine-tuning the network pre-trained on the clean image. In this paper, we propose a novel Dense-Gram Network to more effectively reduce the gap than the conventional strategies and segment degraded images. Extensive experiments demonstrate that the proposed Dense-Gram Network yields state-of-the-art semantic segmentation performance on degraded images synthesized using PASCAL VOC 2012, SUNRGBD, CamVid, and CityScapes datasets. |
3,141 | Pseudomonas oryzagri sp. nov., isolated from a rice field soil | A Gram-stain-negative, aerobic, rod-shaped and non-motile novel bacterial strain, designated MAHUQ-58T, was isolated from soil sample of a rice field. The colonies were observed to be light pink-coloured, smooth, spherical and 0.6-1.0 mm in diameter when grown on nutrient agar (NA) medium for 2 days. Strain MAHUQ-58T was found to be able to grow at 15-40 °C, at pH 5.5-10.0 and with 0-1.0 % NaCl (w/v). Cell growth occurred on tryptone soya agar, Luria-Bertani agar, NA, MacConkey agar and Reasoner's 2A agar. The strain was found to be positive for both oxidase and catalase tests. The strain was positive for hydrolysis of Tween 20 and l-tyrosine. According to the 16S rRNA gene sequence comparisons, the isolate was identified as a member of the genus Pseudomonas and to be closely related to Pseudomonas oryzae WM-3T (98.9 % similarity), Pseudomonas linyingensis LYBRD3-7T (97.7 %), Pseudomonas sagittaria JCM 18195 T (97.6 %) and Pseudomonas guangdongensis SgZ-6T (97.2 %). The novel strain MAHUQ-58T has a draft genome size of 4 536 129 bp (46 contigs), annotated with 4064 protein-coding genes, 60 tRNA genes and four rRNA genes. The average nucleotide identity (ANI) and digital DNA-DNA hybridization (dDDH) values between strain MAHUQ-58T and four closely related type strains were in the range of 85.5-89.5 % and 29.5-38.0 %, respectively. The genomic DNA G+C content was determined to be 67.0 mol%. The predominant isoprenoid quinone was ubiquinone 9. The major fatty acids were identified as C16:0, summed feature 3 (C16 : 1 ω6c and/or C16 : 1 ω7c) and summed feature 8 (C18 : 1 ω6c and/or C18 : 1 ω7c). On the basis of dDDH and ANI values, genotypic results, and chemotaxonomic and physiological data, strain MAHUQ-58T represents a novel species within the genus Pseudomonas, for which the name Pseudomonas oryzagri sp. nov. is proposed, with MAHUQ-58T (=KACC 22005T=CGMCC 1.18518T) as the type strain. |
3,142 | Robust Interactive Image Segmentation Using Convex Active Contours | The state-of-the-art interactive image segmentation algorithms are sensitive to the user inputs and often unable to produce an accurate boundary with a small amount of user interaction. They frequently rely on laborious user editing to refine the segmentation boundary. In this paper, we propose a robust and accurate interactive method based on the recently developed continuous-domain convex active contour model. The proposed method exhibits many desirable properties of an effective interactive image segmentation algorithm, including robustness to user inputs and different initializations, the ability to produce a smooth and accurate boundary contour, and the ability to handle topology changes. Experimental results on a benchmark data set show that the proposed tool is highly effective and outperforms the state-of-the-art interactive image segmentation algorithms. |
3,143 | A Survey of Ensemble Learning: Concepts, Algorithms, Applications, and Prospects | Ensemble learning techniques have achieved state-of-the-art performance in diverse machine learning applications by combining the predictions from two or more base models. This paper presents a concise overview of ensemble learning, covering the three main ensemble methods: bagging, boosting, and stacking, their early development to the recent state-of-the-art algorithms. The study focuses on the widely used ensemble algorithms, including random forest, adaptive boosting (AdaBoost), gradient boosting, extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and categorical boosting (CatBoost). An attempt is made to concisely cover their mathematical and algorithmic representations, which is lacking in the existing literature and would be beneficial to machine learning researchers and practitioners. |
3,144 | Quadrant-based contour features for accelerated shape retrieval system | Shape representation and retrieval are essential research topics of computer vision. This paper proposes a novel feature set to be used in content-based image retrieval systems. The proposed method is an extended version of our previous study which uses contour information of shapes. The previous study calculated the center of mass (CoM) of the shape. By taking the CoM as origin, we created imaginary vectors in every angular direction. From each vector, we extracted three features which are the number of intersections between vector and contour, average distance of intersection points to CoM, and standard deviation of these points. In this method, we extract novel features and decrease the size of the feature set to decrease the computation time. We divide the shape into quadrants and represent each quadrant by nine features. Each shape image is represented by a 4x9 feature vector. We tested the proposed method on MPEG-7 and ETH-80 datasets and compared it with the state-of-art. According to the results, our method decreased the computation time dramatically while giving a state-of-art level retrieval accuracy. |
3,145 | Stochastic Depth Residual Network for Hyperspectral Image Classification | The convolutional neural network (CNN) is a feedforward neural network with deep structure and convolution operation. In the hyperspectral image (HSI) classification, CNN has demonstrated excellent performance in extracting spectral and spatial information. However, the inherent complexity and high dimension of HSIs still limit the performance of most neural network models. The powerful feature extraction ability of CNN is normally achieved by dozens or more layers, which brings a series of problems such as gradient vanishing, overfitting, and slow training speed. In order to address these problems, this article presents a CNN architecture-based stochastic depth residual network (SDRN), which is specially designed for HSI data. This model takes the original 3-D cube as the input and 3-D convolution is used to extract abundant spectral and spatial features through corresponding residual blocks. In order to reduce the training time, we adopt a stochastic depth strategy. For each small batch, a sublayer is randomly discarded by an identity function. During the testing stage, the residual network with complete depth is used. Experiments on three datasets and a comparison of the state-of-art methods show that SDRN has great advantages in accuracy and training time compared with state-of-the-art HSI classification methods. |
3,146 | Phase separation of the microtubule-associated protein tau | The aggregation and misfolding of the neuronal microtubule-associated protein tau is closely linked to the pathology of Alzheimer's disease and several other neurodegenerative diseases. Recent evidence suggest that tau undergoes liquid-liquid phase separation in vitro and forms or associates with membrane-less organelles in cells. Biomolecular condensation driven by phase separation can influence the biological activities of tau including its ability to polymerize tubulin into microtubules. In addition, the high concentrations that tau can reach in biomolecular condensates provide a mechanism to promote its aggregation and the formation of amyloid fibrils potentially contributing to the pathology of different tauopathies. Here, the authors discuss the role of tau phase separation in physiology and disease. |
3,147 | Advances in Electrical Machine, Power Electronic, and Drive Condition Monitoring and Fault Detection: State of the Art | Recently, research concerning electrical machines and drives condition monitoring and fault diagnosis has experienced extraordinarily dynamic activity. The increasing importance of these energy conversion devices and their widespread use in uncountable applications have motivated significant research efforts. This paper presents an analysis of the state of the art in this field. The analyzed contributions were published in most relevant journals and magazines or presented in either specific conferences in the area or more broadly scoped events. |
3,148 | Shadow-Consistent Semi-Supervised Learning for Prostate Ultrasound Segmentation | Prostate segmentation in transrectal ultrasound (TRUS) image is an essential prerequisite for many prostate-related clinical procedures, which, however, is also a long-standing problem due to the challenges caused by the low image quality and shadow artifacts. In this paper, we propose a Shadow-consistent Semi-supervised Learning (SCO-SSL) method with two novel mechanisms, namely shadow augmentation (Shadow-AUG) and shadow dropout (Shadow-DROP), to tackle this challenging problem. Specifically, Shadow-AUG enriches training samples by adding simulated shadow artifacts to the images to make the network robust to the shadow patterns. Shadow-DROP enforces the segmentation network to infer the prostate boundary using the neighboring shadow-free pixels. Extensive experiments are conducted on two large clinical datasets (a public dataset containing 1,761 TRUS volumes and an in-house dataset containing 662 TRUS volumes). In the fully-supervised setting, a vanilla U-Net equipped with our Shadow-AUG&Shadow-DROP outperforms the state-of-the-arts with statistical significance. In the semi-supervised setting, even with only 20% labeled training data, our SCO-SSL method still achieves highly competitive performance, suggesting great clinical value in relieving the labor of data annotation. Source code is released at https://github.com/DIAL-RPI/SCO-SSL. |
3,149 | The Effects of Iconicity and Conventionalization on Word Order Preferences | Of the six possible orderings of the three main constituents of language (subject, verb, and object), two-SOV and SVO-are predominant cross-linguistically. Previous research using the silent gesture paradigm in which hearing participants produce or respond to gestures without speech has shown that different factors such as reversibility, salience, and animacy can affect the preferences for different orders. Here, we test whether participants' preferences for orders that are conditioned on the semantics of the event change depending on (i) the iconicity of individual gestural elements and (ii) the prior knowledge of a conventional lexicon. Our findings demonstrate the same preference for semantically conditioned word order found in previous studies, specifically that SOV and SVO are preferred differentially for different types of events. We do not find that iconicity of individual gestures affects participants' ordering preferences; however, we do find that learning a lexicon leads to a stronger preference for SVO-like orders overall. Finally, we compare our findings from English speakers, using an SVO-dominant language, with data from speakers of an SOV-dominant language, Turkish. We find that, while learning a lexicon leads to an increase in SVO preference for both sets of participants, this effect is mediated by language background and event type, suggesting that an interplay of factors together determines preferences for different ordering patterns. Taken together, our results support a view of word order as a gradient phenomenon responding to multiple biases. |
3,150 | Hydroquinone derivatives attenuate biofilm formation and virulence factor production in Vibrio spp | Gram-negative Vibrio parahaemolyticus is a halophilic human pathogen known to be the leading cause of food poisoning associated with consuming uncooked or undercooked seafood. The increasing presence and contamination of seafood have caused serious safety concerns in food facilities. Notably, it can form biofilms on food surfaces that confer resistance to antimicrobial treatments. Therefore, in the present study, the antibacterial, antibiofilm, and antivirulence activities of hydroquinone (HQ) and its 16 derivatives were investigated against V. parahaemolyticus and V. harveyi. Representative active antibacterial and antibiofilm compounds, 2,3-dimethylhydroquinone (2,3-DMHQ) and 2,5-ditert-butylhydroquinone (DBHQ), were further examined using a crystal violet assay, biochemical reactions, live cell imaging, and scanning electron microscopy. 2,3-DMHQ with a minimum inhibitory concentration (MIC) of 20 μg/mL completely inhibited biofilm formation at a sub-MIC of 15 μg/mL. And, DBHQ with an MIC of ˃1000 μg/mL reduced biofilm formation by 70 % at sub-MIC of 25 μg/mL. Both 2,3-DMHQ and DBHQ inhibited protease and indole production as well as motility phenotypes. 2,3-DMHQ decreased fimbriae production and hydrophobicity whereas DBHQ did not. Transcriptomic studies revealed that genes related to biofilm, quorum sensing (QS), and hemolysin were downregulated. In addition, 2,3-DMHQ and DBHQ prevented biofilm formation of V. parahaemolyticus on squid surfaces and 2,3-DMHQ reduced the presence of V. parahaemolyticus in a boiled shrimp model. Toxicity assays using the Caenorhabditis elegans and seed germinations models showed that they were non-to-mildly toxic. These results suggest that 2,3-DMHQ and DBHQ possess the antimicrobial properties required to control V. parahaemolyticus planktonic and biofilm states in food production facilities. |
3,151 | Moving object segmentation in Daubechies complex wavelet domain | Motion segmentation is a crucial step in video analysis and is associated with a number of computer vision applications. This paper introduces a new method for segmentation of moving object which is based on double change detection technique applied on Daubechies complex wavelet coefficients of three consecutive frames. Daubechies complex wavelet transform for segmentation of moving object has been chosen as it is approximate shift invariant and has a better directional selectivity as compared to real valued wavelet transform. Double change detection technique is used to obtain video object plane by inter-frame difference of three consecutive frames. Double change detection technique also provides automatic detection of appearance of new objects. The proposed method does not require any other parameter except Daubechies complex wavelet coefficients. Results of the proposed method for segmentation of moving objects are compared with results of other state-of-the-art methods in terms of visual performance and a number of quantitative performance metrics viz. Misclassification Penalty, Relative Foreground Area Measure, Pixel Classification Based Measure, Normalized Absolute Error, and Percentage of Correct Classification. The proposed method is found to have high degree of segmentation accuracy than the other state-of-the-art methods. |
3,152 | Fast Superpixel Based Subspace Low Rank Learning Method for Hyperspectral Denoising | Sequential data, such as video frames and event data, have been widely applied in the real-world. As a special kind of sequential data, hyperspectral images (HSIs) can be regarded as a sequence of 2-D images in the spectral dimension, which can be effectively utilized for distinguishing different land-covers according to the spectral sequences. This paper presents a novel noise reduction method based on superpixel-based subspace low rank representation for hyperspectral imagery. First, under the framework of a linear mixture model, the original hyperspectral cube is assumed to be low rank in the spectral domain, which could be represented by decomposing HSI data into two sub-matrices of lower ranks. Meanwhile, due to the high correlation of neighboring pixels, the spectra within each neighborhood would also promote low rankness, and the local spatial low rankness could be exploited by enforcing the nuclear norm within superpixel-based regions in the decomposed subspace. The superpixels are easily and effectively generated by utilizing state-of-the-art superpixel segmentation algorithms in the first principal component of the original HSI. Moreover, benefiting from the subspace decomposition, the proposed method has an overwhelming superiority in computational cost than the state-of-the-art LR-based methods. The final model could be efficiently solved by the augmented Lagrangian method. Experimental results on simulated and real hyperspectral data sets validate that the proposed method produces superior performance than other state-of-the-art denoising methods in terms of quantitative assessment and visual quality. |
3,153 | Dilated Convolutional Pixels Affinity Network for Weakly Supervised Semantic Segmentation | This paper studies semantic segmentation primarily under image-level weak-supervision. Most state-of-the-art technologies have recently used deep classification networks to create small and sparse discriminatory seed regions of each interest target as pseudo-labels for training segmentation networks, which achieve inferior performance compared with the fully supervised setting. We propose a Dilated convolutional pixels affinity network (DCPAN) to localize and expand the seed regions of objects to bridge this gap. Although introduced dilated convolutional units enable capture of additional location information of objects, it falsely highlighted true negative regions as dilated rate enlarge. To address this problem, we properly integrate dilated convolutional units with different dilated rates and self-attention mechanisms to obtain pixel affinity measure matrix for promoting classification network to generate high-quality object seed regions as pseudo-labels; thus, the performance of the segmentation network is boosted. Furthermore, although our approach seems simple, our method obtains a competitive performance, and experiments show that the performance of DCPAN outperforms other state-of-art approaches in weakly-supervised settings, which only use image-level labels on the Pascal VOC 2012 dataset. |
3,154 | Changes in temporomandibular joint morphology in class II patients treated with fixed mandibular repositioning and evaluated through 3D imaging: a systematic review | To estimate the effects of skeletal class II malocclusion treatment using fixed mandibular repositioning appliances on the position and morphology of the temporomandibular joint (TMJ). Two independent reviewers performed comprehensive electronic searches of MEDLINE, EMBASE, EBM reviews and Scopus (until May 5, 2015). The references of the identified articles were also manually searched. All studies investigating morphological changes of the TMJ articular disc, condyle and glenoid fossa with 3D imaging following non-surgical fixed mandibular repositioning appliances in growing individuals with class II malocclusions were included in the analysis. Of the 269 articles initially reviewed, only 12 articles used magnetic resonance imaging and two articles used computed tomography (CT) or cone-beam CT images. Treatment effect on condyle and glenoid fossa was discussed in eight articles. Treatment effect on TMJ articular disc position and morphology was discussed in seven articles. All articles showed a high risk of bias due to deficient methodology: inadequate consideration of confounding variables, blinding of image assessment, selection or absence of control group and outcome measurement. Reported changes in osseous remodelling, condylar and disc position were contradictory. The selected articles failed to establish conclusive evidence of the exact nature of TMJ tissue response to fixed mandibular repositioning appliances. |
3,155 | Deinococcus betulae sp. nov. and Deinococcus arboris sp. nov., novel bacteria isolated from bark of birch tree ( Betula platyphylla) | Two reddish-coloured bacterial strains (HMF7604T and HMF7620T) were isolated from bark of birch tree (Betula platyphylla) together with two strains (designed as HMF7603 and HMF7618). Cells were observed to be Gram-stain-negative, oval- to short rod-shaped and non-motile. Phylogenetic analysis based on 16S rRNA gene sequences indicated that the four isolates belonged to the genus Deinococcus, family Deinococcaceae. They had the highest similarities (95.4-95.6 %) to Deinococcus multiflagellatus ID1504T, with which they formed a clade in phylogenetic trees. Menaquinone-8 was the only respiratory quinone. The predominant fatty acids were summed feature 3 (C16 : 1 ω7c and/or C16 : 1 ω6c), C15 : 1 ω6c, C17 : 0 and C16 : 0. Strain HMF7604T contained two unidentified phosphoglycolipids, nine unidentified glycolipids, one unidentified aminolipid, three unidentified phospholipids and three unidentified polar lipids, while strain HMF7620T contained one unidentified phosphoglycolipid, four unidentified glycolipids, one unidentified aminophospholipid, one unidentified phospholipid and one unidentified polar lipid. The DNA G+C contents of strains HMF7604T and HMF7620T were 65.6 and 65.7 mol%, respectively. The average nucleotide identity and digital DNA-DNA hybridization values between the two isolates and their close relative D. multiflagellatus were 81.1-95.3 and 24.5-61.6 %, respectively. Based on the results of phenotypic and phylogenetic characterizations, the four isolates are considered to represent two novel species of the genus Deinococcus, for which the names Deinococcus betulae sp. nov. and Deinococcus arboris sp. nov. are proposed. The type strains are HMF7604T (=KCTC 43354T=NBRC 115489T) and HMF7620T (=KCTC 43051T=NBRC 113959T). |
3,156 | Evidence and Impacts of Nanoplastic Accumulation on Crop Grains | Nanoplastics are emerging pollutants of global concern. Agricultural soil is becoming a primary sink for nanoplastics generated from plastic debris. The uptake and accumulation of nanoplastics by crops contaminate the food chain and pose unexpected risks to human health. However, whether nanoplastics can enter grains and their impact on the grains of crop grown in contaminated soil is still unknown. Here, the translocation of polystyrene nanoplastics (PS-NPs) in crops, including peanut (Arachis hypogaea L.) and rice (Oryza sativa L.) is investigated. It is demonstrated PS-NPs translocation from the root and accumulation in the grains at the maturation stage. The treatment with PS-NPs (250 mg kg-1 ) increases the empty-shell numbers of rice grain by 35.45%, thereby decreasing the seed-setting rate of rice by 3.02%, and also decreases the average seed weight of peanuts by 3.45%. Moreover, PS-NPs exerted adverse effects on nutritional quality, such as decreasing the content of mineral elements, amino acids, and unsaturated fatty acids. To the knowledge, this is the first report of the presence of nanoplastics in the grains of crop plants grown in soil containing nanoplastics, and the results highlight the impact of nanoplastics on the yield and nutritional quality of crop grains. |
3,157 | Yellow-Green Luminescence Due to Polarity-Dependent Incorporation of Carbon Impurities in Self-Assembled GaN Microdisk | Yellow-green luminescence (YGL) competes with near-bandgap emission (NBE) for carrier recombination channels, thereby reducing device efficiency; yet uncovering the origin of YGL remains a major challenge. In this paper, nearly stress-free and low dislocation density self-assembled GaN microdisks were synthesized by Na-flux method. The YGL of GaN microdisks highly depend on their polar facets. Variable accelerating voltage/power CL, variable temperature PL, and Raman spectroscopy are further performed to clarify the origin of polarity dependence of GaN microdisk YGL behavior, which indicates its independence of dislocations, surface effects, stress, crystalline quality, and gallium vacancies. It was found that the incorporation ability of carbon impurities in the polar (0001) facet is greater than that in the semipolar (101̅1) facets, producing higher content of CN or CNON defects, resulting in a more pronounced YGL in the polar (0001) facet of GaN. |
3,158 | Multimodal Single-Cell Translation and Alignment with Semi-Supervised Learning | Single-cell multi-omics technologies enable comprehensive interrogation of cellular regulation, yet most single-cell assays measure only one type of activity-such as transcription, chromatin accessibility, DNA methylation, or 3D chromatin architecture-for each cell. To enable a multimodal view for individual cells, we propose Polarbear, a semi-supervised machine learning framework that facilitates missing modality profile prediction and single-cell cross-modality alignment. Polarbear learns to translate between modalities by using data from co-assay measurements coupled with the large quantity of single-assay data available in public databases. This semi-supervised scheme mitigates issues related to low cell quantities and high sparsity in co-assay data. Polarbear first pre-trains a beta-variational autoencoder for each modality using both co-assay and single-assay profiles to learn robust representations of individual cells, and it then uses the co-assay labels to train a translator between these cell representations. This semi-supervised framework enables us to predict missing modality profiles and match single cells across modalities with improved accuracy compared with fully supervised methods, thus facilitating multimodal data integration. |
3,159 | Deep structured residual encoder-decoder network with a novel loss function for nuclei segmentation of kidney and breast histopathology images | To improve the process of diagnosis and treatment of cancer disease, automatic segmentation of haematoxylin and eosin (H & E) stained cell nuclei from histopathology images is the first step in digital pathology. The proposed deep structured residual encoder-decoder network (DSREDN) focuses on two aspects: first, it effectively utilized residual connections throughout the network and provides a wide and deep encoder-decoder path, which results to capture relevant context and more localized features. Second, vanished boundary of detected nuclei is addressed by proposing an efficient loss function that better train our proposed model and reduces the false prediction which is undesirable especially in healthcare applications. The proposed architecture experimented on three different publicly available H&E stained histopathological datasets namely: (I) Kidney (RCC) (II) Triple Negative Breast Cancer (TNBC) (III) MoNuSeg-2018. We have considered F1-score, Aggregated Jaccard Index (AJI), the total number of parameters, and FLOPs (Floating point operations), which are mostly preferred performance measure metrics for comparison of nuclei segmentation. The evaluated score of nuclei segmentation indicated that the proposed architecture achieved a considerable margin over five state-of-the-art deep learning models on three different histopathology datasets. Visual segmentation results show that the proposed DSREDN model accurately segment the nuclear regions than those of the state-of-the-art methods. |
3,160 | Understanding Performance Limitations of Cu(In,Ga)Se-2 Solar Cells due to Metastable Defects-A Route toward Higher Efficiencies | Thin-film Cu(In,Ga)Se-2 solar cells reach power conversion efficiencies exceeding 23% and nonradiative recombination in the bulk is reported to limit device performance. The diode factor has not received much attention, although it limits the fill factor, and therefore the efficiency, for state-of-the-art solar cells. Herein, the diode factor of Cu(In,Ga)Se-2 absorbers, measured by photoluminescence spectroscopy, and of solar cells, measured by current-voltage and capacitance-voltage characteristics, are compared, supported by simulations using rate equations of generation and recombination. It is found that the diode factor is already increased in the neutral zone of the absorber due to metastable defects, such as the V-Se-V-Cu defect found in Cu(In,Ga)Se-2, because of an increased net acceptor density upon minority-carrier injection. The metastable and persistent increase of the net acceptor density has a detrimental effect on the device performance. Diode factors of 1 and efficiencies exceeding 24% are expected when, in current state-of-the-art Cu(In,Ga)Se-2 solar cells, the formation of metastable defects is suppressed. |
3,161 | Visibility enhancement of fog degraded images using adaptive defogging function | In the field of image processing, analyzing fog-affected images is challenging, as their visibility is degraded. In the absence of state-of-the-art image processing techniques to mitigate the impact of high-density fog, an adaptive-function-based image-defogging technique is proposed in this paper. The proposed technique accurately enhances such degraded images by adjusting the contrast and brightness based on a suitable threshold operator. The images are subsequently characterized as foggy or non-foggy on the basis of objective evaluation. The experimental results have proven that the proposed method achieves superior performance in terms of qualitative evaluation on non-reference metric (i.e., in terms of e = 0.468, sigma = 0, r = 1.8857) and reference metric (i.e. in terms of MSE = 1580, PSNR = 19.2126, NCC = 0.4873, SC = 0.4684, MD = 60, NAE = 0.2229) compared with nine state-of-the-art dehazing methods. Furthermore, based on the average computational time achieved by the proposed method (0.36 s using a test set of 2000 images), it can be highly suitable for real-time applications. |
3,162 | 3D facial expression recognition using kernel methods on Riemannian manifold | Automatic human Facial Expressions Recognition (FER) is becoming of increased interest. FER finds its applications in many emerging areas such as affective computing and intelligent human computer interaction. Most of the existing work on FER has been done using 2D data which suffers from inherent problems of illumination changes and pose variations. With the development of 3D image capturing technologies, the acquisition of 3D data is becoming a more feasible task. The 3D data brings a more effective solution in addressing the issues raised by its 2D counterpart. State-of-the-art 3D FER methods are often based on a single descriptor which may fail to handle the large inter-class and intra-class variability of the human facial expressions. In this work, we explore, for the first time, the usage of covariance matrices of descriptors, instead of the descriptors themselves, in 3D FER. Since covariance matrices are elements of the non-linear manifold of Symmetric Positive Definite (SPD) matrices, we particularly look at the application of manifold-based classification to the problem of 3D FER. We evaluate the performance of the proposed framework on the BU-3DFE and the Bosphorus datasets, and demonstrate its superiority compared to the state-of-the-art methods. (C) 2017 Elsevier Ltd. All rights reserved. |
3,163 | Laypeople's perceptions of the effects of event repetition, reporting delay, and emotion on children's and adults' memory | For crimes such as child abuse and family violence, jurors' assessments of memory reports from key witnesses are vital to case outcomes in court. Since jurors are not experts on memory, the present research measured laypeople's (i.e., non-experts') beliefs about how three key factors affect witnesses' memory reports for an experienced event: how frequently an event was experienced (repeated, single), the delay between experiencing and reporting the event, and the emotional valence of the event. Across two studies, lay participants completed an online survey that measured their beliefs about each factor. In Study 1, 51 participants completed a survey about how the three factors affect children's memory. In Study 2, another 51 participants completed a survey about how the three factors affect adult's memory. Across both studies, delays were believed to worsen memory, and emotion was believed to improve memory. Beliefs about single and repeated events showed different patterns across the studies. In Study 1, participants' beliefs about children's memory for repeated experience were variable. In Study 2, participants believed that adults' memory was worse for repeated events than single events. Overall, laypeople demonstrated many accurate beliefs about memory, but showed some confusion about children's memory for repeated events. |
3,164 | Mexicanolide limonoids from the seeds of Khaya ivorensis with antimicrobial activity | The methanol extract of the seeds of Khaya ivorensis afforded two new mexicanolide limonoids, ivorensines A and B (1 and 2), together with one known compound, ruageanin D (3). The structures of the isolated compounds were established based on 1 D and 2 D (1H-1H COSY, HMQC, and HMBC) NMR spectroscopy, in addition to high resolution mass spectrometry. The isolated limonoids were tested in vitro for antimicrobial potentials against 5 pathogenic microorganisms. As a result, compounds 1-3 exhibited antimicrobial activity against the tested Gram negative bacteria at the minimum inhibitory concentration values less than 50 μg/ml. |
3,165 | Intriguing H2S Tolerance of the PtRu Alloy for Hydrogen Oxidation Catalysis in PEMFCs: Weakened Pt-S Binding with Slower Adsorption Kinetics | High quality of hydrogen is the key to the long lifetime of proton-exchange membrane fuel cell (PEMFC) vehicles, while trace H2S impurities in hydrogen significantly affect their durability and fuel expense. Herein, we demonstrate a robust PtRu alloy catalyst with an intriguing H2S tolerance as the PEMFC anode, showing a stronger antipoisoning capability toward hydrogen oxidation reaction compared with the Pt/C anode. The PtRu/C-based single PEMFC shows approximately 14.3% loss of cell voltage after 3 h operation with 1 ppm of H2S in hydrogen, significantly lower than that of Pt/C-based PEMFCs (65%). By adopting PtRu/C as the anode, the H2S limit in hydrogen can be increased to 1.7 times that of the Pt/C anode, assuming that the PEMFC runs for 5000 h, which is conductive for the cost reduction of hydrogen purification. The three-electrode electrochemical test indicates that PtRu/C exhibits a slower adsorption kinetics toward S2- species with poisoning rates of 0.02782, 0.02982, and 0.03682 min-1 at temperatures of 25, 35, and 45 °C, respectively, all lower than those of Pt/C. X-ray absorption fine structure spectra indicate the weakened Pt-S binding for PtRu/C in comparison to Pt/C with a longer Pt-S bond length. Density functional theory calculation analyses reveal that adsorption energy of sulfur on the Pt surface was reduced for PtRu/C, showing 1-10% decrease at different Pt sites for (111), (110), and (100) planes, which is ascribed to the downshifted Pt d-band center caused by the ligand and strain effects due to the introduction of second metallic Ru. This work provides a valuable guide for the development of the H2S-tolerant catalysts for long-term application of PEMFCs. |
3,166 | Iron deficiency and soil-transmitted helminth infection: classic and neglected connections | Beyond participating in the oxygen transport by red blood cells, iron is an essential micronutrient and contributes to different physiological pathways and processes, such as cell proliferation, DNA repair, and other homeostatic functions. Iron deficiency affects millions of people, especially children and pregnant women. The consequences of iron deficiency are diverse, including inadequate child development, impaired cognition, and reduced productivity. Several factors contribute to iron deficiency, such as iron-poor diet, genetic factors, and infection with soil-transmitted helminths (STHs), especially roundworms (Ascaris lumbricoides), hookworms (Necator americanus and Ancylostoma duodenale), and whipworms (Trichuris trichiura). This review updates and summarizes the role of STHs as drivers of iron deficiency. Also, the poorly explored connections between STH infection, geophagia (a pica manifestation), immune response, and iron deficiency are discussed, highlighting how iron deficiency may act as a risk factor for infections by STHs, in addition to being a consequence of intestinal parasitic infections. Finally, strategies for control and management of iron deficiency and STH infection are described. |
3,167 | Inhibitor; An Uncommon But Vexing Challenge In North Indian Patients With Hemophilia A | Factor VIII replacement is the mainstay of treatment in hemophilia A but may lead to the development of inhibitors. While a vexing clinical problem, some observations suggest that the presence of inhibitors may not necessarily portend a higher bleeding risk. Our aim was to assess the prevalence and clinicopathological correlates of inhibitors in a well characterized cohort of Indian patients with HA patients. We retrospectively reviewed the clinical details and laboratory findings of consecutive hemophilia A patients attending a north-Indian tertiary-care center from 2010 to 2020. Among 592 patients with HA, inhibitors were detected in 35 patients (5.9%). Prevalence of inhibitors in moderate and severe hemophilia was 4.2% and 6.7%, respectively. Most patients with inhibitors had history of transfusion with factor VIII alone (54.3%) or a combination of factor VIII concentrate and other blood-products (42.9%). Intracranial bleed was significantly more frequent in patients with inhibitors compared to those without inhibitors (20% vs. 4.1%; p-0.001). Time dependent and immediately acting inhibitors were seen in 60% and 40% patients, respectively. High-titre (> 5 BU) and low-titre inhibitors (< 5 BU) were detected in 28 (80%) and 7 (20%) patients, respectively. Prevalence of inhibitors in our cohort was 5.9% and most had high-titre, time dependent inhibitors. These patients may have a higher risk of intracranial bleeding. |
3,168 | The effect of a dam on the copper accumulation in estuarine sediment and associated nematodes in a Mekong estuary | Dam construction across the main flow of an estuary can greatly contribute to a high accumulation of inorganic contaminants. However, it remains unknown to what extend externally available heavy metals are incorporated into biota living in those contaminated environments. In this study, the heavy metal copper was investigated both in the sediment and in the tissues of nematodes taken from the subtidal zone in the Ba Lai estuary where a dam is present, and compared with samples from the dam-free Ham Luong estuary, both part of the Mekong Delta. Samples were taken in the dry season of 2017 in four stations in the Ba Lai estuary with two stations in the downstream part from the dam and two upstream. Similar locations with respect to the distance were sampled in the dam-free estuary. The internal copper concentration in nematodes was measured by applying micro X-ray fluorescence. The results showed that both internal and sediment copper concentrations were different between the two estuaries and among estuarine sections. The highest copper concentration in nematodes was found in the upstream section of Ba Lai estuary where the greatest accumulation of sedimentary copper was observed, while the dammed downstream part was lowest in internal copper accumulation. Moreover, there was more variation in the copper levels between the two sections within the dammed estuary compared to those in Ham Luong. These observations might point to the contribution of the Ba Lai dam to the increase of copper contaminants in the benthic environment leading to accumulation in nematodes. |
3,169 | Extending morphological covariance | Mathematical morphology conversely to linear image analysis approaches specialises in capturing the spatial relations among pixels. This inherent potential has been exploited in the context of texture characterisation, with granulometry along with morphological covariance being the two main tools of the morphological arsenal for this task However, with the advent of new and powerful texture analysis approaches in the last years (e.g. local binary patterns, MR8), they have been left relatively behind the state-of-the-art, in the light of the present challenges of this field, particularly illumination, rotation and scale invariant characterisation. In this paper, we present a set of extensions for morphological covariance, inspired from differential morphological profiles, that enhance its rotation and illumination invariance capacity. The proposed approach is tested extensively against the state-of-the-art, using the Outex, CUReT, KTH-TIPS, KTH-TIPS2 and ALOT databases, where it exhibits either a superior or comparable performance. (C) 2012 Elsevier Ltd. All rights reserved. |
3,170 | GeoTrackNet--A Maritime Anomaly Detector Using Probabilistic Neural Network Representation of AIS Tracks and A Contrario Detection | Representing maritime traffic patterns and detecting anomalies from them are key to vessel monitoring and maritime situational awareness. We propose a novel approach--referred to as GeoTrackNet--for maritime anomaly detection from AIS data streams. Our model exploits state-of-the-art neural network schemes to learn a probabilistic representation of AIS tracks and a contrario detection to detect abnormal events. The neural network provides a new means to capture complex and heterogeneous patterns in vessels' behaviours, while the a contrario detector takes into account the fact that the learnt distribution may be location-dependent. Experiments on a real AIS dataset comprising more than 4.2 million AIS messages demonstrate the relevance of the proposed method compared with state-of-the-art schemes. |
3,171 | Using complex network analysis for water quality assessment in large water distribution systems | Assessing and modelling the water quality in a water distribution system (WDS) are highly important to ensure a reliable supply with sufficient water quality. Owing to the high computational burden of such an analysis, frequently, simplifications are required or surrogate models are used (e.g., reducing the level of detail of the network model), neglecting significant aspects. For large (currently all-pipe) models and/or recurrent simulations (e.g., integrated studies, sensitivity analysis, deep uncertainty analysis, design, and optimization), the computational burden further increases. In this study, a novel complex network analysis-based approach for high-computational efficiency water quality assessment in a WDS is developed and comprehensively tested (R2 values in comparison with state-of-the-art nodal water qualities in median of 0.95 are achieved). The proposed model is successfully utilized in a design study to identify the design solutions exceeding water quality thresholds with a correct identification rate between 96% and 100%. The computational efficiency is determined to be a factor 4.2e-06 less than that of state-of-the-art models. Therefore, the proposed model significantly improves the water quality assessment for such tasks in large WDSs. |
3,172 | Public and occupational health risks related to lead exposure updated according to present-day blood lead levels | Lead is an environmental hazard that should be addressed worldwide. Over time, human lead exposure in the western world has decreased drastically to levels comparable to those among humans living in the preindustrial era, who were mainly exposed to natural sources of lead. To re-evaluate the potential health risks associated with present-day lead exposure, a two-pronged approach was applied. First, recently published population metrics describing the adverse health effects associated with lead exposure at the population level were critically assessed. Next, the key results of the Study for Promotion of Health in Recycling Lead (SPHERL; NCT02243904) were summarized and put in perspective with those of the published population metrics. To our knowledge, SPHERL is the first prospective study that accounted for interindividual variability between people with respect to their vulnerability to the toxic effects of lead exposure by assessing the participants' health status before and after occupational lead exposure. The overall conclusion of this comprehensive review is that mainstream ideas about the public and occupational health risks related to lead exposure urgently need to be updated because a large portion of the available literature became obsolete given the sharp decrease in exposure levels over the past 40 years. |
3,173 | Crowd Counting in Low-Resolution Crowded Scenes Using Region-Based Deep Convolutional Neural Networks | Crowd counting and density estimation is an important and challenging problem in the visual analysis of the crowd. Most of the existing approaches use regression on density maps for the crowd count from a single image. However, these methods cannot localize individual pedestrian and therefore cannot estimate the actual distribution of pedestrians in the environment. On the other hand, detection-based methods detect and localize pedestrians in the scene, but the performance of these methods degrades when applied in high-density situations. To overcome the limitations of pedestrian detectors, we proposed a motion-guided filter (MGF) that exploits spatial and temporal information between consecutive frames of the video to recover missed detections. Our framework is based on the deep convolution neural network (DCNN) for crowd counting in the low-to-medium density videos. We employ various state-of-the-art network architectures, namely, Visual Geometry Group (VGG16), Zeiler and Fergus (ZF), and VGGM in the framework of a region-based DCNN for detecting pedestrians. After pedestrian detection, the proposed motion guided filter is employed. We evaluate the performance of our approach on three publicly available datasets. The experimental results demonstrate the effectiveness of our approach, which significantly improves the performance of the state-of-the-art detectors. |
3,174 | Dominant Color Extraction with K-Means for Camera Characterization in Cultural Heritage Documentation | The camera characterization procedure has been recognized as a convenient methodology to correct color recordings in cultural heritage documentation and preservation tasks. Instead of using a whole color checker as a training sample set, in this paper, we introduce a novel framework named the Patch Adaptive Selection with K-Means (P-ASK) to extract a subset of dominant colors from a digital image and automatically identify their corresponding chips in the color chart used as characterizing colorimetric reference. We tested the methodology on a set of rock art painting images captured with a number of digital cameras. The characterization approach based on the P-ASK framework allows the reduction of the training sample size and a better color adjustment to the chromatic range of the input scene. In addition, the computing time required for model training is less than in the regular approach with all color chips, and obtained average color differences Delta E-ab(*) lower than two CIELAB units. Furthermore, the graphic and numeric results obtained for the characterized images are encouraging and confirms that the P-ASK framework based on the K-means algorithm is suitable for automatic patch selection for camera characterization purposes. |
3,175 | Isolation and characterization of Newcastle disease virus from biological fluids using column chromatography | Newcastle disease virus (NDV), belonging to the species avian orthoavulavirus 1, genus Orthoavulavirus, and family Paramyxoviridae, is responsible for Newcastle disease in poultry and other avian species. It has shown significant potential as an oncolytic virus and as a vector for vaccine delivery. NDV from infected biological serum is usually isolated or purified using density gradient ultracentrifugation. However, it has many disadvantages, including the fact that it is time consuming and can process only a limited quantity of sample at one time. In our study, native agarose gel electrophoresis and dynamic light scattering (DLS) analysis showed that NDV carried a net negative surface charge. Thus, we purified the virus using a HiTrap Q Sepharose Fast Flow anion exchange column with salt elution. Hemagglutination assay and plaque assay showed that the procedure yielded high-purity NDV particles with a recovery of more than 80%, and the process was fast and simple. The purity of the virus was confirmed using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot analysis. The hydrodynamic volume and 'dry state' diameter of the purified NDV were analyzed using dynamic light scattering and transmission electron microscopy and were to be in the range of 200-300 nm. The viruses did not exhibit any deviation from their known physical properties. The genome of the virus was also detected by amplifying a 423-bp region using reverse transcription-polymerase chain reaction. Our study confirmed that NDV could be effectively purified using an anion exchange column. In addition, the procedure could be easily upscaled or downscaled based on the experimental requirements. |
3,176 | Townes-Brocks syndrome with craniosynostosis in two siblings | This report describes a novel truncating c.709C > T p.(Gln237*) SALL1 variant in two siblings exhibiting sagittal craniosynostosis as a unique feature of Townes-Brocks syndrome (TBS, OMIM #107480). TBS is a rare autosomal dominant syndrome with variable phenotypes, including anorectal, renal, limb, and ear abnormalities, which results from heterozygous variants in the SALL1 gene, predominantly located in the 802 bp "hot spot region" within exon 2. Recent studies have suggested that aberrations in primary cilia and sonic hedgehog signalling contribute to the TBS phenotypes. The presence of the novel c.709C > T p.(Gln237*) SALL1 variant was confirmed in both the siblings and their father, whereas no mutations currently associated with craniosynostosis were detected. We hypothesise that the truncating c.709C > T p.(Gln237*) SALL1 variant, which occurs outside the "hot spot region" and inside the glutamine-rich domain coding region, could interfere with ciliary signalling and mechanotransduction, contributing to premature fusion of calvarial sutures. This report broadens the genetic and phenotypic spectrum of TBS and provides the first clinical evidence of craniosynostosis as a novel feature of the syndrome. |
3,177 | Geometry-Based Demosaicking | Demosaicking is a particular case of interpolation problems where, from a scalar image in which each pixel has either the red, the green or the blue component, we want to interpolate the full-color image. State-of-the-art demosaicking algorithms perform interpolation along edges, but these edges are estimated locally. We propose a level-set-based geometric method to estimate image edges, inspired by the image inpainting literature. This method has a time complexity of O(S), where S is the number of pixels in the image, and compares favorably with the state-of-the-art algorithms both visually and in most relevant image quality measures. |
3,178 | Discrete crow-inspired algorithms for traveling salesman problem | Crow search algorithm is one of bio-inspired optimization algorithms which is essentially derived for solving continuous based optimization problems. Although many main-frame discrete optimizers are available, they still have some performance challenges. This paper proposes three discrete crow inspired algorithms for enhancing the performance of the original crow search algorithm when it is applied for solving discrete traveling salesman problems. The proposed algorithms are derived based on modular arithmetic, basic operators and dissimilar solutions techniques. Each technique guarantees switching from continuous spaces into discrete spaces without losing information. Such algorithms are called Modular Arithmetic, Basic Operators, and Dissimilar Solutions algorithms. For evaluating their performance, the proposed algorithms are compared with the most state-of-the-art discrete optimizers for solving 111 instances of traveling salesman problems. Simulation results illustrate that, the performance of the proposed algorithms is much better than the performance of most state-of-the-art discrete optimizers in terms of the average optimal solutions accuracy, the average errors from the optimal solutions and the average of computational time. |
3,179 | The future of the Affective Neuroscience Personality Scales: A reflection on seven pressing matters | The Affective Neuroscience Personality Scales (ANPS) were designed to provide researchers in the mental sciences with an inventory to assess primary emotional systems according to Pankseppian Affective Neuroscience Theory (ANT). The original ANPS, providing researchers with such a tool, was published in 2003. In the present brief communication, about 20 years later, we reflect upon some pressing matters regarding the further development of the ANPS. We touch upon problems related to disentangling traits and states of the primary emotional systems with the currently available versions of the ANPS and upon its psychometric properties and its length. We reflect also on problems such as the large overlap between the SADNESS and FEAR dimensions, the disentangling of PANIC and GRIEF in the context of SADNESS, and the absence of a LUST scale. Lastly, we want to encourage scientists with the present brief communication to engage in further biological validation of the ANPS. |
3,180 | Longer-Term Adverse Effects of Selenate Exposures on Hematological and Serum Biochemical Variables in Air-Breathing Fish Channa punctata (Bloch, 1973) and Non-air Breathing Fish Ctenopharyngodon Idella (Cuvier, 1844): an Integrated Biomarker Response Approach | To examine the spectrum of selenium toxicity between hardy and less hardy species of the same life stages, short-term and longer-term exposures in juvenile air-breathing fish Channa punctata (Bloch, 1973) and non-air-breathing fish Ctenopharyngodon idella (Cuvier, 1844) were assessed. Acute exposures revealed a greater 96-h median lethal concentration (LC50) for C. punctata (14.67 mg/l) compared to C. idella (7.98 mg/l). During their chronic exposure, both fishes' hemoglobin content (Hb), red blood cells (RBC), and hematocrit (HCT) markedly decreased (p < 0.05), although their clotting time (CT) significantly increased. At 96 h, immune-modulation was observed where total protein and serum globulin levels in both fishes considerably decreased (p < 0.05) compared to the first exposure at 0 days, although total glucose, triglyceride, cholesterol, and albumin levels in both fishes significantly increased (p < 0.05) at 30 days. The lower cholesterol levels in C. punctata compared to C. idella are suggestive of a disrupted cholesterol transformation pathway. The greater total protein, triglyceride, albumin, and globulin levels in C. punctata compared to C. idella are suggestive of a comparatively robust immune capacity. In essence, selenium toxicity in the wild could manifest as disrupted metabolic pathways and downregulated immune capacity for less hardy species. In general, both fish species displayed significant alterations in their hematological and biochemical responses with increased exposure duration and elevated toxicant concentrations. This comparative investigation could improve the knowledge-spectrum of selenium toxicity in the wild as well as an understanding of secondary stress responses critically evident in hematological and biochemical parameters. |
3,181 | Accidental and Late Diagnosis of Type A Aortic Dissection: Mimicking Unstable Angina Pectoris | Aortic dissection is an infrequent diagnosis that usually presents with acute onset of sharp and severe tearing pain. It rarely presents with atypical symptoms, accompanied by a higher mortality risk that arises the delay in diagnosis. In this report, we discuss a type A aortic dissection case with a presentation of heaviness-like chest pain with no evidence of aortic dissection in his first echocardiography. The patient was treated for acute coronary syndrome (ACS), but on the follow-up, echocardiography aortic dissection was diagnosed accidentally. Differentiation between ACS and aortic dissection is critical in patient management. Each one has an entirely different treatment approach, and misdiagnosis can lead to catastrophic outcomes. |
3,182 | First Occurrence of Megastigmane Glucosides in a Plant of Retama Genus | Chemical investigation of Retama sphaerocarpa collected in Algeria resulted in the isolation of two megastigmane glucosides, compounds 1 and 2, along with a series of isoflavones and phenol derivatives. Compound 1, named retamoside, was new and its structure was determined by extensive application of spectroscopic methods, including HRMS, 1D and 2D NMR and CD. The anti-inflammatory properties of co-occurring main megastigmane, saurobaccioside B (2) and structurally related vomifoliol (3) on LPS-stimulated murine macrophages RAW 274.7 have been evaluated. |
3,183 | Failure analysis requirements for nanoelectronics | Failure analysis (FA) plays a vital role in the development and manufacture of integrated circuits. However, instrumental limits are already threatening FA in the tenth-micron CMOS realm, and nanoelectronic devices will find key analytical tools two orders of magnitude removed in capability. This paper will introduce state-of-the-art microelectronic failure analysis processes, instrumentation, and principles. It will discuss the major limitations and future prospects determined from industry roadmaps. Specifically highlighted is the need for a fault isolation methodology for failure analysis of fully integrated nanoelectronics; devices. |
3,184 | The Brakeless co-regulator can directly activate and repress transcription in early Drosophila embryos | The Brakeless protein performs many important functions during Drosophila development, but how it controls gene expression is poorly understood. We previously showed that Brakeless can function as a transcriptional co-repressor. In this work, we perform transcriptional profiling of brakeless mutant embryos. Unexpectedly, the majority of affected genes are down-regulated in brakeless mutants. We demonstrate that genomic regions in close proximity to some of these genes are occupied by Brakeless, that over-expression of Brakeless causes a reciprocal effect on expression of these genes, and that Brakeless remains an activator of the genes upon fusion to an activation domain. Together, our results show that Brakeless can both repress and activate gene expression. A yeast two-hybrid screen identified the Mediator complex subunit Med19 as interacting with an evolutionarily conserved part of Brakeless. Both down- and up-regulated Brakeless target genes are also affected in Med19-depleted embryos, but only down-regulated targets are influenced in embryos depleted of both Brakeless and Med19. Our data provide support for a Brakeless activator function that regulates transcription by interacting with Med19. We conclude that the transcriptional co-regulator Brakeless can either activate or repress transcription depending on context. |
3,185 | Predicting CT Image From MRI Data Through Feature Matching With Learned Nonlinear Local Descriptors | Attenuation correction for positron-emission tomography (PET)/magnetic resonance(MR) hybrid imaging systems and dose planning for MR-based radiation therapy remain challenging due to insufficient high-energy photon attenuation information. We present a novel approach that uses the learned nonlinear local descriptors and feature matching to predict pseudo computed tomography (pCT) images from T1-weighted and T2-weighted magnetic resonance imaging (MRI) data. The nonlinear local descriptors are obtained by projecting the linear descriptors into the nonlinear high-dimensional space using an explicit feature map and low-rank approximation with supervised manifold regularization. The nearest neighbors of each local descriptor in the input MR images are searched in a constrained spatial range of the MR images among the training dataset. Then the pCT patches are estimated through-nearest neighbor regression. The proposed method for pCT prediction is quantitatively analyzed on a dataset consisting of paired brain MRI and CT images from 13 subjects. Our method generates pCT images with a mean absolute error (MAE) of 75.25 +/- 18.05 Hounsfield units, a peak signal-to-noise ratio of 30.87 +/- 1.15 dB, a relative MAE of 1.56 +/- 0.5% in PET attenuation correction, and a dose relative structure volume difference of 0.055 +/- 0.107% in D-98%, as compared with true CT. The experimental results also show that our method outperforms four state-of-the-art methods. |
3,186 | The application of clustering analysis for the critical areas on TFT-LCD panel | For thin film transistor-liquid crystal displays (TFT-LCD) factories in Taiwan, yield performance had become as an important competitiveness determinant during the competitive environment. As we known, the market for LCDs has grown at over 20% on average per annum and the downward pricing trend had also promoted LCD applications. However, only few studies were proposed to address the related issues for process analysis in TFT-LCD industry from the viewpoint of systems. Particularly, the defect status (i.e. abnormal position) on TFT-LCD panel may represent the clustering effect when there are many defect counts on it. Hence, performing the clustering analysis for those abnormal positions will be an important issue to be addressed in TFT-LCD process. In this study, we will propose an approach incorporating fuzzy adaptive resonance theory (Fuzzy ART) and stepwise regression techniques to achieve such process analysis. Besides, an illustrative case owing to TFT-LCD manufacturer at Tainan Science Park in Taiwan will be applied to verifying the rationality and feasibility of our proposed procedure. (c) 2006 Elsevier Ltd. All rights reserved. |
3,187 | Next generation grid storage "HYDRAstor" | HYDRAstor is a grid storage product developed in accordance with the NEC's three REAL IT PLATFORM concepts of "Flexibility," "Security" and "Comfort," in order to resolve serious issues surrounding storage in recent years. This paper introduces the state-of-the-art grid storage architecture of HYDRAstor and introduces its core technologies, Dynamic Topology, DataRedux and Distributed Resilient Data. |
3,188 | Mobility scaling in short-channel length strained Ge-on-insulator P-MOSFETs | The hole transport characteristics in partially strained (0.5%) Ge p-channel MOSFETs formed on silicon-germanium-on-insulator (SGOI) substrates were investigated for gate lengths down to 65 min. We demonstrate that high hole mobility is maintained down to the shortest channel lengths. The channel conductance from these devices is measured and compared to state-of-the-art high-performance Si channel P-MOSFETs. |
3,189 | Novel convolutional neural network architecture for improved pulmonary nodule classification on computed tomography | Computed tomography (CT) is widely used to locate pulmonary nodules for preliminary diagnosis of the lung cancer. However, due to high visual similarities between malignant (cancer) and benign (non-cancer) nodules, distinguishing malignant from malign nodules is not an easy task for a thoracic radiologist. In this paper, a novel convolutional neural network (ConvNet) architecture is proposed to classify the pulmonary nodules as either benign or malignant. Due to the high variance of nodule characteristics in CT scans, such as size and shape, a multi-path, multi-scale architecture is proposed and applied in the proposed ConvNet to improve the classification performance. The multi-scale method utilizes filters with different sizes to more effectively extracted nodule features from local regions, and the multi-path architecture combines features extracted from different ConvNet layers thereby enhancing the nodule features with respect to global regions. The proposed ConvNet is trained and evaluated on the LUNGx Challenge database, and achieves a sensitivity of 0.887 and a specificity of 0.924 with an area under the curve (AUC) of 0.948. The proposed ConvNet achieves a 14% AUC improvement compared to the state-of-the-art unsupervised learning approach. The proposed ConvNet also outperforms the other state-of-the-art ConvNets explicitly designed for pulmonary nodule classification. For clinical usage, the proposed ConvNet could potentially assist the radiologists to make diagnostic decisions in CT screening. |
3,190 | Learning 3D Head Pose From Synthetic Data: A Semi-Supervised Approach | Accurate head pose estimation from 2D image data is an essential component of applications such as driver monitoring systems, virtual reality technology, and human-computer interaction. It enables a better determination of user engagement and attentiveness. The most accurate head pose estimators are based on Deep Neural Networks that are trained with the supervised approach and rely primarily on the accuracy of training data. The acquisition of real head pose data with a wide variation of yaw, pitch and roll is a challenging task. Publicly available head pose datasets have limitations with respect to size, resolution, annotation accuracy and diversity. In this work, a methodology is proposed to generate pixel-perfect synthetic 2D headshot images rendered from high-quality 3D synthetic facial models with accurate head pose annotations. A diverse range of variations in age, race, and gender are also provided. The resulting dataset includes more than 300k pairs of RGB images with corresponding head pose annotations. A wide range of variations in pose, illumination and background are included. The dataset is evaluated by training a state-of-the-art head pose estimation model and testing against the popular evaluation-dataset Biwi. The results show that training with purely synthetic data generated using the proposed methodology achieves close to state-of-the-art results on head pose estimation which are originally trained on real human facial datasets. As there is a domain gap between the synthetic images and real-world images in the feature space, initial experimental results fall short of the current state-of-the-art. To reduce the domain gap, a semi-supervised visual domain adaptation approach is proposed, which simultaneously trains with the labelled synthetic data and the unlabeled real data. When domain adaptation is applied, a significant improvement in model performance is achieved. Additionally, by applying a data fusion-based transfer learning approach, better results are achieved than previously published work on this topic. |
3,191 | Image fusion method based on the advection equation | We present an innovative method based on the linear advection equation, an important partial differential equation, to perform the fusion of images. The basic idea of this method is to insert the relevant information from other source images into the current source image through an advection process. Furthermore, we present the discrete scheme of this model and compare it with classical fusion approaches, the diffusion equation-based method, and some state-of-the-art fusion approaches on three groups of fusion images that are often used in the image fusion research. The results of experiments show that the fusion method based on the advection equation is comparable with the best of the classical, diffusion-based, and state-of-the-art methods. The high "weighted performance metric" Q(AB/F) of fused images certifies that the relevant information is well injected from the input to the output images. Moreover, this method has fewer adjustable parameters with settings that affect the metric Q(AB/F) less than other methods, and the evolution from input to output is also faster than the diffusion-based method. In addition, this model allows us to cope with noisy source image fusion by adding a diffusion term in the equation, thereby combining the denoising process with the fusion process. (C) 2014 SPIE and IS&T |
3,192 | Investigating the effects of demand flexibility on electricity retailers' business through a tri-level optimisation model | The investigation of the effects of demand flexibility on the pricing strategies and the profits of electricity retailers has recently emerged as a highly interesting research area. However, the state-of-the-art, bi-level optimisation modelling approach makes the unrealistic assumption that retailers treat wholesale market prices as exogenous, fixed parameters. This study proposes a tri-level optimisation model, which drops this assumption and represents the wholesale market clearing process endogenously, thus capturing the realistic implications of a retailer's pricing strategies and the resulting demand response on the wholesale market prices. The scope of the examined case studies is three-fold. First of all, they demonstrate the interactions between the retailer, the flexible consumers and the wholesale market and analyse the fundamental effects of the consumers' time-shifting flexibility on the retailer's revenue from the consumers, its cost in the wholesale market and its overall profit. Furthermore, they analyse how these effects of demand flexibility depend on the retailer's relative size in the market and the strictness of the regulatory framework. Finally, they highlight the added value of the proposed tri-level model by comparing its outcomes against the state-of-the-art bi-level modelling approach. |
3,193 | Promoting microbiology education through the iGEM synthetic biology competition | Synthetic biology has developed rapidly in the 21st century. It covers a range of scientific disciplines that incorporate principles from engineering to take advantage of and improve biological systems, often applied to specific problems. Methods important in this subject area include the systematic design and testing of biological systems and, here, we describe how synthetic biology projects frequently develop microbiology skills and education. Synthetic biology research has huge potential in biotechnology and medicine, which brings important ethical and moral issues to address, offering learning opportunities about the wider impact of microbiological research. Synthetic biology projects have developed into wide-ranging training and educational experiences through iGEM, the International Genetically Engineered Machines competition. Elements of the competition are judged against specific criteria and teams can win medals and prizes across several categories. Collaboration is an important element of iGEM, and all DNA constructs synthesized by iGEM teams are made available to all researchers through the Registry for Standard Biological Parts. An overview of microbiological developments in the iGEM competition is provided. This review is targeted at educators that focus on microbiology and synthetic biology, but will also be of value to undergraduate and postgraduate students with an interest in this exciting subject area. |
3,194 | Low-Complexity Linear Zero-Forcing for the MIMO Broadcast Channel | Maximizing the sum capacity in the multiple-input multiple-output (MIMO) broadcast channel requires the use of dirty paper coding (DPC). However, practical implementations of DPC which are nearly optimum exhibit high computational complexity. As an alternative to DPC linear zero-forcing can be used where the multiuser interference is completely canceled by linear beamforming. Determining the optimum user allocation, transmit and receive filters thereby constitutes a combinatorial and nonconvex optimization problem. To circumvent its direct solution and therefore reduce complexity, we propose an algorithm that successively allocates data streams to users and, in contrast to state-of-the-art approaches, includes the receive filters into the optimization. We then show several steps that reduce the complexity of the algorithm at marginal performance losses. Thus, performance of state-of-the-art approaches can be maintained while the computational complexity is reduced considerably, as it is shown by a detailed complexity analysis and simulation results. |
3,195 | Delay-dependent forgetting in object recognition and object location test is dependent on strain and test | The object recognition and object location task (ORT and OLT, respectively) have been applied in preclinical research to evaluate the effects of treatments on memory. Although both tasks look quite similar, they differ with respect to the brain structures involved in the memory performance. The characterization of the memory performance in both tasks is important to understand treatment effects. Since there are no previous studies that compared strain differences in delay-dependent forgetting in both tasks, Wistar and Long Evans rats were tested in both the ORT and the OLT at different intervals. The data showed that in the ORT the delay-dependent forgetting was similar for Wistar and Long Evans rats. However, the forgetting curve was different for both strains in the OLT: the Long Evans rats the forgetting took a longer interval. This study indicates that delay-dependent forgetting in the ORT and OLT is strain and test dependent. It is suggested that before testing treatments the forgetting curve of a specific strain should be tested in this type of tasks. |
3,196 | Research on Open Practice Teaching of Off-Campus Art Appreciation Based on ICT | Art appreciation is an effective way to promote artistic literacy and is also an important component of aesthetic education in school. With the help of information and communication technology, the authors organized open practice teaching for students to learn art appreciation outside school. During the COVID-19 epidemic, local art appreciation education could not be carried out in the city where the authors' school is located. With the support of mobile positioning technology and information platforms, students were able to carry out 32 art appreciation activities in their hometowns during this period. Through the mobile positioning information submitted by students, feedback questionnaires, after-view data, and other data, learning achievements were identified. A correlation analysis of the data submitted by the students on the information platform confirmed that satisfaction with the art appreciation activity directly affected their interest in art. The correlation reached 0.78. Satisfaction was strongly correlated with psychological expectations (0.67) and art information obtained in the early stage (0.61). The authors propose that using information and communication technology to carry out art appreciation education outside the school is the way to promote the sustainable development of aesthetic education in school. |
3,197 | Predict, Share, and Recycle Your Way to Low-power Nanophotonic Networks | High static power consumption is widely regarded as one of the largest bottlenecks in creating scalable optical NoCs. The standard techniques to reduce static power are based on sharing optical channels and modulating the laser. We show in this article that state-of-the-art techniques in these areas are suboptimal, and there is a significant room for further improvement. We propose two novel techniques-a neural network-based method for laser modulation by predicting optical traffic and a distributed and altruistic algorithm for channel sharing-that are significantly closer to a theoretically ideal scheme. In spite of this, a lot of laser power still gets wasted. We propose to reuse this energy to heat micro-ring resonators (achieve thermal tuning) by efficiently recirculating it. These three methods help us significantly reduce the energy requirements. Our design consumes 4.7x lower laser power as compared to other state-of-the-art proposals. In addition, it results in a 31% improvement in performance and 39% reduction in ED2 for a suite of Splash2 and Parsec benchmarks. |
3,198 | Enhanced Microcystis Aeruginosa removal and novel flocculation mechanisms using a novel continuous co-coagulation flotation (CCF) | Co-coagulation flotation (CCF) is a novel flotation technology that renders more efficient algal removal compared to traditional mechanical coagulation flotation (MCF) due to a short residence time (< 30 s) and fast rising behavior of algal flocs (> 250 m·h-1). This study compared the algal removal performance using continuous CCF and MCF using water samples taken from Lake Dianchi with severe Microcystis aeruginosa blooms. Removal efficiency, dosage of coagulant/flocculant, rising velocity and structural characteristics of the resulting flocs in the two processes were systematically compared. The results show that CCF could save >50 % polyaluminum chloride (PAC) and polyacrylamide (PAM) compared with MCF when the removal efficiency was both over 95 %. The average rising velocity of flocs in CCF could reach 254.3 m·h-1, much higher than that in MCF (154.5 m·h-1). In the respective optimal coagulation conditions, the flocs formed in CCF (G = 164.8 s-1) were larger (1843 ± 128 μm) and more spherical with a higher fractal dimension (Df = 1.85 ± 0.01) than those generated in MCF (G = 34.1 s-1). The Stokes's Law was found to correctly predict the rising velocity of spherical flocs with large fractal dimensions (Df > 1.7). In contrast, the Haarhoff and Edzwald's extended equation was more suitable for calculating the rising velocity of irregular flocs with small fractal dimension. This study provides new insights into the mechanisms of the enhanced algal removal by CCF and lays foundation for developing cost-efficient algal mitigation processes. |
3,199 | Cryptanalysis of MD2 | This paper considers the hash function MD2 which was developed by Ron Rivest in 1989. Despite its age, MD2 has withstood cryptanalytic attacks until recently. This paper contains the state-of-the-art cryptanalytic results on MD2, in particular collision and preimage attacks on the full hash function, the latter having complexity 2(73), which should be compared to a brute-force attack of complexity 2(128). |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.