Search

Exposure to particulate matter with a diameter

Background Patients with heart failure (HF) experience continuous changes in symptom distress, care needs, social support, and meaning in life from acute decompensation to chronic phases. The longitudinal relationship between these four factors and quality of life (QOL) was not fully explored. Aims To simultaneously investigate the relationship between all factors and QOL from hospitalization to 6 months after discharge, and the impact of the changes in these factors on QOL at different time points. Methods A longitudinal design with panel research (4 time points) was used. From January 2017 to December 2019, patients hospitalized due to acute decompensated HF were consecutively enrolled and followed up for 6 months. Patients were interviewed with questionnaires assessing symptom distress, care needs, social support, meaning in life and QOL at hospitalization and 1, 3 and 6 months after discharge. Results A total of 184 patients completed 6 months of follow-up. From baseline to 6 months, QOL continuously improved along with decreases in symptoms and care needs, but increases in social support and meaning in life. Better QOL was associated with younger age, higher education level, economic independence, less symptom distress and care needs, and stronger meaning in life (p p p p Conclusions Although symptom distress is associated with QOL after acute decompensated HF, QOL cannot be improved only by improvement in symptoms. With differential duration of improvement in each factor, the integration of alleviation in care needs and strengthening in social support and meaning in life might provide additional benefits in QOL.

Background Mismatch negativity (MMN) reflects the functional integrity of sensory memory function. With the advantages of independence of individual's focused attention and behavioral cooperation, this neurophysiological signal is particularly suitable for investigating elderly with cognitive decline such as amnestic mild cognitive impairment (aMCI). However, the existing results remain substantially inconsistent whether these patients show deficits of MMN. In order to reconcile the previous disputes, the present study used magnetoencephalography combined with distributed source imaging methods to determine the source‐level magnetic mismatch negativity (MMNm) in aMCI. Methods A total of 26 healthy controls (HC) and 26 patients with aMCI underwent an auditory oddball paradigm during the MEG recordings. MMNm amplitudes and latencies in the bilateral superior temporal gyrus, inferior frontal gyrus, and inferior parietal lobule (IPL) were compared between HC and aMCI groups. The correlations of MMNm responses with performance of auditory/verbal memory tests were examined. Finally, MMNm and its combination with verbal/auditory memory tests were submitted to receiver operating characteristic (ROC) curve analysis. Results Compared to HC, patients with aMCI showed significantly delayed MMNm latencies in the IPL. Among the patients with aMCI, longer MMNm latencies of left IPL were associated with lower scores of Chinese Version Verbal Learning Test (CVVLT). The ROC curve analysis revealed that the combination of MMNm latencies of left IPL and CVVLT scores yielded a moderate accuracy in the discrimination of aMCI from HC at an individual level. Conclusions Our data suggest dysfunctional MMNm in patients with aMCI, particularly in the IPL., This study uses MEG to compare source‐level magnetic mismatch negativity (MMNm) in healthy elderly and patients with amnestic mild cognitive impairment (aMCI). Patients with aMCI show MMNm prolongation in the inferior parietal lobule, and such a deficit correlates with poorer verbal learning test.

BackgroundEnhanced recovery pathways can be further improved for postoperative sore throat (POST) which usually occurs after surgery with general anesthesia. Medications have shown some effectiveness in treating and preventing POST, but acupuncture or related techniques with better safety and less cost likely can be used as an alternative or adjuvant therapy to treat perioperative symptoms by stimulating acupuncture point (acupoint). Therefore, we aim to conduct a meta-analysis to assess whether acupoint stimulation help patients prevent or treat POST in adults undergoing tracheal intubation for general anesthesia.MethodsPublication in PubMed, the Cochrane Central Register, ScienceDirect, and ClinicalTrial.gov were surveyed from Jan. 2000 through Jan. 2020. Studies that compared intervention between point stimulation and none or sham point stimulation, were included. Primary outcomes were the incidence and severity of POST at 24h. Secondary outcomes were the incidence of postoperative nausea and vomiting, choking cough, and sputum.ResultsThree randomized control trials and one comparative study involving 1358 participants were included. Compared with control, acupoint stimulation was associated with a reduced incidence (risk ratio, 0.3; 95% confidence interval (CI), 0.2–0.45; p < 0.001) and severity (standardized mean difference, −2.21; 95% CI, −2.67 to −1.76; p < 0.001) of POST. Secondary outcomes are also in favor of acupoint stimulation. There were no significant adverse events related to acupoint stimulation. Subgroup, the sensitivity, and the trial sequence analyses confirmed that the finding for POST was adequate.ConclusionsAcupoint stimulation with various methods may reduce the occurrence of POST. It could be considered as one of nonpharmacological ways to prevent POST in enhanced recovery pathways. Further rigorous studies are needed to determine the effectiveness of acupoint stimulation.QuestionCan acupoint stimulation prevent postoperative sore throat after tracheal intubation?FindingsAcupoint stimulation by acupuncture or related techniques more significantly reduces the incidence and the severity of postoperative sore throat than non- /sham-treatment at 24 hours.MeaningAcupoint stimulation by acupuncture or related techniques could be an effective, nonpharmacological approach to prevent postoperative sore throat in enhanced recovery after tracheal intubation.

Introduction Acute respiratory distress syndrome (ARDS) due to coronavirus disease 2019 (COVID-19) often leads to mortality. Outcomes of patients with COVID-19-related ARDS compared to ARDS unrelated to COVID-19 is not well characterized. Areas covered We performed a systematic review of PubMed, Scopus, and MedRxiv 11/1/2019 to 3/1/2021, including studies comparing outcomes in COVID-19-related ARDS (COVID-19 group) and ARDS unrelated to COVID-19 (ARDS group). Outcomes investigated were duration of mechanical ventilation-free days, intensive care unit (ICU) length-of-stay (LOS), hospital LOS, and mortality. Random effects models were fit for each outcome measure. Effect sizes were reported as pooled median differences of medians (MDMs), mean differences (MDs), or odds ratios (ORs). Expert opinion Ten studies with 2,281 patients met inclusion criteria (COVID-19: 861 [37.7%], ARDS: 1420 [62.3%]). There were no significant differences between the COVID-19 and ARDS groups for median number of mechanical ventilator-free days (MDM: −7.0 [95% CI: −14.8; 0.7], p = 0.075), ICU LOS (MD: 3.1 [95% CI: −5.9; 12.1], p = 0.501), hospital LOS (MD: 2.5 [95% CI: −5.6; 10.7], p = 0.542), or all-cause mortality (OR: 1.25 [95% CI: 0.78; 1.99], p = 0.361). Compared to the general ARDS population, results did not suggest worse outcomes in COVID-19-related ARDS.

Summary: Background: The aim of this study was to assess the safety of several modified Nuss procedures for severe pectus excavatum (PE). Methods: Thirty-four patients with severe PE underwent the Nuss procedure: 10 underwent slanting-directed bar insertion (group A); 11 underwent standard Nuss procedure (group B); and 13 underwent Nuss procedure with subxiphoid assistance (group C). All the patients met the criteria of having a Haller index greater than 4.5, assessed from chest computed tomography. Besides, the transverse length of the most depressed point and the 2-intercostal left slant length between the heart and the anterior chest wall were measured. Results: All patients were followed up for 6–45 months (mean 31.4 ± 11.38 months). None of the patients suffered from injuries to the pericardium, heart or lungs. There were no significant differences in age, Haller’s index, operation time and postoperative stay among the three groups. However, two patients in group B experienced bar rotation and subsequently required reoperation vs the other two groups (p

Outdoors organization is necessary in order to learn from prior experiences so as to create a proactive visitor risk management approach. Using retrospective analysis, this paper presents the epide...

Zeroth-order (ZO) optimization is a subset of gradient-free optimization that emerges in many signal processing and machine learning (ML) applications. It is used for solving optimization problems similarly to gradient-based methods. However, it does not require the gradient, using only function evaluations. Specifically, ZO optimization iteratively performs three major steps: gradient estimation, descent direction computation, and the solution update. In this article, we provide a comprehensive review of ZO optimization, with an emphasis on showing the underlying intuition, optimization principles, and recent advances in convergence analysis. Moreover, we demonstrate promising applications of ZO optimization, such as evaluating robustness and generating explanations from black-box deep learning (DL) models and efficient online sensor management.

Coronavirus disease 2019 (COVID-19), which is caused by infection with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has become a global pandemic. It is unclear whether convalescing patients have a risk of reinfection. We generated a rhesus macaque model of SARS-CoV-2 infection that was characterized by interstitial pneumonia and systemic viral dissemination mainly in the respiratory and gastrointestinal tracts. Rhesus macaques reinfected with the identical SARS-CoV-2 strain during the early recovery phase of the initial SARS-CoV-2 infection did not show detectable viral dissemination, clinical manifestations of viral disease, or histopathological changes. Comparing the humoral and cellular immunity between primary infection and rechallenge revealed notably enhanced neutralizing antibody and immune responses. Our results suggest that primary SARS-CoV-2 exposure protects against subsequent reinfection in rhesus macaques.

Background: The rise in human papillomavirus (HPV) infection rates over the last few decades in the USA has contributed to a significant increase in the overall incidence of patients diagnosed with squamous cell carcinoma of the head and neck. These head and neck carcinomas develop in the oropharynx, with more than 90% of them caused by infection with high-risk HPV type 16. Patients diagnosed with HPV-induced oropharyngeal squamous cell carcinomas (OPSCCs) have a better prognosis and treatment response than those diagnosed with head and neck cancers caused by alcohol consumption and tobacco use. To identify patients with HPV-positive OPSCC, new guidelines recommend positive staining of oropharyngeal tissues for p16 INK4a (p16) by immunohistochemistry (IHC). Herein we discuss the testing algorithm that was adopted to address discrepant results between p16 IHC and a DNA in situ hybridization (ISH) test used routinely to diagnose HPV-positive OPSCC patients. Methods: A DNA polymerase chain reaction (PCR) test that amplifies HPV16 and HPV18 E7 was developed to aid in the diagnosis of HPV-positive OPSCC in a subset of patients. Specimens from these patients stained positive for p16 by an IHC test, but negative for high-risk HPV by a commercial DNA ISH test. Moreover, these results did not match the histopathological characteristics of the specimens, nor the clinical presentations of the patients. Results: Of 21 patients' specimens that were tested for p16 by IHC, 11 specimens showed concordant results with the high-risk HPV 16/18 DNA ISH test. Whereas, in eight p16 IHC positive specimens, HPV viral DNA was not detected by HPV16/18 DNA ISH, and two specimens were not tested by DNA ISH. When these eight p16 IHC positive specimens with discrepant p16 IHC and DNA ISH results were further tested by DNA PCR, six specimens showed concordance with p16 IHC with positive results for HPV16 E7, while two specimens were negative for HPV16 E7 by DNA PCR. All tested specimens were negative for HPV18 E7 by DNA PCR. Thus, the addition of the HPV16 and HPV18 E7 DNA PCR test identified a significant number of false negative test results by the HPV16/18 DNA ISH test and likely several false positive results by p16 IHC. Conclusions: Inclusion of an HPV16 E7 DNA PCR test improved the robustness of HPV-associated OPSCC diagnosis in patients with discrepant results from p16 IHC staining and a DNA ISH test, and identified patients for proper management with less misclassification. World J Oncol. 2020;11(1):1-8 doi: https://doi.org/10.14740/wjon1243 Â

The state-of-the-art machine learning approaches are based on classical von Neumann computing architectures and have been widely used in many industrial and academic domains. With the recent development of quantum computing, researchers and tech-giants have attempted new quantum circuits for machine learning tasks. However, the existing quantum computing platforms are hard to simulate classical deep learning models or problems because of the intractability of deep quantum circuits. Thus, it is necessary to design feasible quantum algorithms for quantum machine learning for noisy intermediate scale quantum (NISQ) devices. This work explores variational quantum circuits for deep reinforcement learning. Specifically, we reshape classical deep reinforcement learning algorithms like experience replay and target network into a representation of variational quantum circuits. Moreover, we use a quantum information encoding scheme to reduce the number of model parameters compared to classical neural networks. To the best of our knowledge, this work is the first proof-of-principle demonstration of variational quantum circuits to approximate the deep $Q$-value function for decision-making and policy-selection reinforcement learning with experience replay and target network. Besides, our variational quantum circuits can be deployed in many near-term NISQ machines., Accepted for publication by IEEE Access

Due to their crucial role in many decision-making tasks, Deep Neural Networks (DNNs) are common targets for a large array of integrity breaches. In this paper, we propose AID, a novel methodology to Attest the Integrity of DNNs. AID generates a set of test cases called edge-points that can reveal whether a model has been compromised. AID does not require access to parameters of the DNN and can work with a restricted black-box access to the model, which makes it applicable to most real life scenarios. Experimental results show that AID is highly effective and reliable. With at most four edge-points, AID is able to detect eight representative integrity breaches including backdoor, poisoning, and compression attacks, with zero false-positive.

Adversarial learning methods and their applications such as generative adversarial network, adversarial robustness, and security and privacy, have prevailed and revolutionized the research in machine learning and data mining. Their importance has not only been emphasized by the research community but also been widely recognized by the industry and the general public. Continuing the synergies in previous years, this third annual workshop aims to advance this research field. The AdvML'21 workshop consists of three tracks: (i) open-call paper submissions; (ii) invited speakers; and (iii) rising star awards and presentations. The full details about the workshop can be found at https://sites.google.com/view/advml.

As the application of deep neural networks proliferates in numerous areas such as medical imaging, video surveillance, and self driving cars, the need for explaining the decisions of these models has become a hot research topic, both at the global and local level. Locally, most explanation methods have focused on identifying relevance of features, limiting the types of explanations possible. In this paper, we investigate a new direction by leveraging latent features to generate contrastive explanations; predictions are explained not only by highlighting aspects that are in themselves sufficient to justify the classification, but also by new aspects which if added will change the classification. The key contribution of this paper lies in how we add features to rich data in a formal yet humanly interpretable way that leads to meaningful results. Our new definition of "addition" uses latent features to move beyond the limitations of previous explanations and resolve an open question laid out in Dhurandhar, et. al. (2018), which creates local contrastive explanations but is limited to simple datasets such as grayscale images. The strength of our approach in creating intuitive explanations that are also quantitatively superior to other methods is demonstrated on three diverse image datasets (skin lesions, faces, and fashion apparel). A user study with 200 participants further exemplifies the benefits of contrastive information, which can be viewed as complementary to other state-of-the-art interpretability methods., Accepted to KDD 2021

With the diversified sources satisfying users' needs, many online service platforms collect information from multiple sources in order to provide a set of useful information to the users. However, existing recommendation systems are mostly designed for single-source data, and thus fail to recommend multi-source review packages since the interplay between the reviews of different sources is not properly modeled. In fact, modeling the interplay between different sources is challenging because 1) two reviews may conflict with each other, 2) different users have different preferences on review sources, and 3) users' preferences to each source may change under different scenarios. To address these challenges, we propose Self-Attentive Recommendation for multi-source review Package (SARP), for predicting how useful the user feels to the package, while simultaneously reflecting how much the user is affected by each review. Specifically, SARP jointly considers the relationships of every user, purpose, and review source to learn better latent representations. A self-attention module is further used for integrating source representations and the review ratings, following a multi-layer perceptron (MLP) for the prediction tasks. Experimental results on the self-constructed dataset and public dataset demonstrate that the proposed model outperforms the state-of-the-art approaches.

Preschool children have a higher respiratory rate per unit body weight than adults, and their respiratory systems are not mature. Hence, children may have more health risks associated with ...

As we enter the Internet of Things (IoT) era in which the communication network is becoming increasingly dynamic, heterogeneous, and complex, it is desirable to have cognitive communication systems and networks that possess multiple interacting capabilities for situation assessment, resource management, online/distributed learning, big-data processing, and intelligent decision making. AI techniques, such as deep learning, probabilistic graph model, and reinforcement learning, aided with big data and IoT, provide a wide variety of tools and solutions to many new problems encountered in the design, operation, and optimization of cognitive communication systems and networking, including resource management, situation assessment, channel identification, anomaly detection, root cause analysis, and online/distributed learning.

Given the accelerating adoption of apps and mobile technology in the hospitality and tourism industry, it is critical to understand customer experience of mobile apps as emerging marketing ...

We propose a Paired Few-shot GAN (PFS-GAN) model for learning generators with sufficient source data and a few target data. While generative model learning typically needs large-scale training data, our PFS-GAN not only uses the concept of few-shot learning but also domain shift to transfer the knowledge across domains, which alleviates the issue of obtaining low-quality generator when only trained with target domain data. The cross-domain datasets are assumed to have two properties: (1) each target-domain sample has its source-domain correspondence and (2) two domains share similar content information but different appearance. Our PFS-GAN aims to learn the disentangled representation from images, which composed of domain-invariant content features and domain-specific appearance features. Furthermore, a relation loss is introduced on the content features while shifting the appearance features to increase the structural diversity. Extensive experiments show that our method has better quantitative and qualitative results on the generated target-domain data with higher diversity in comparison to several baselines., accepted in ICASSP 2021

Diffuse alveolar hemorrhage (DAH) in systemic lupus erythematosus (SLE) is associated with significant mortality, requiring a thorough understanding of its complex mechanisms to develop novel therapeutics for disease control. Activated p53-dependent apoptosis with dysregulated long non-coding RNA (lncRNA) expression is involved in the SLE pathogenesis and correlated with clinical activity. We examined the expression of apoptosis-related p53-dependent lncRNA, including H19, HOTAIR and lincRNA-p21 in SLE-associated DAH patients. Increased lincRNA-p21 levels were detected in circulating mononuclear cells, mainly in CD4+ and CD14+ cells. Higher expression of p53, lincRNA-p21 and cell apoptosis was identified in lung tissues. Lentivirus-based short hairpin RNA (shRNA)-transduced stable transfectants were created for examining the targeting efficacy in lncRNA. Under pristane stimulation, alveolar epithelial cells had increased p53, lincRNA-p21 and downstream Bax levels with elevated apoptotic ratios. After pristane injection, C57/BL6 mice developed DAH with increased pulmonary expression of p53, lincRNA-p21 and cell apoptosis. Intra-pulmonary delivery of shRNA targeting lincRNA-p21 reduced hemorrhage frequencies and improved anemia status through decreasing Bax expression and cell apoptosis. Our findings demonstrate increased p53-dependent lncRNA expression with accelerated cell apoptosis in the lungs of SLE-associated DAH patients, and show the therapeutic potential of targeting intra-pulmonary lncRNA expression in a pristane-induced model of DAH.

Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) has spread across the world and impacted global healthcare systems. For clinical patients, COVID-19 not only induces pulmonary lesions but also affects extrapulmonary organs. An ideal animal model that mimics COVID-19 in humans in terms of the induced systematic lesions is urgently needed. Here, we report that Syrian hamster is highly permissive to SARS-CoV-2 and exhibit diffuse alveolar damage and induced extrapulmonary multi-organs damage, including spleen, lymph nodes, different segments of alimentary tract, kidney, adrenal gland, ovary, vesicular gland and prostate damage, at 3–7 days post inoculation (dpi), based on qRT-PCR,in situhybridization and immunohistochemistry detection. Notably, the adrenal gland is a novel target organ, with abundant viral RNA and antigen expression detected, accompanied by focal to diffuse inflammation. Additionally, viral RNA was also detected in the corpus luteum of the ovary, vesicular gland and prostate. Focal lesions in liver, gallbladder, myocardium, and lymph nodes were still present at 18 dpi, suggesting potential damage after disease. Our findings illustrate systemic histological observations and the viral RNA and antigen distribution in infected hamsters during disease and convalescence to recapitulate those observed in humans with COVID-19, providing helpful data to the pathophysiologic characterization of SARS-CoV-2-induced systemic disease and the development of effective treatment strategies.

Influenza A virus may circulate simultaneously with the SARS-CoV-2 virus, leading to more serious respiratory diseases during this winter. However, the influence of these viruses on disease outcome when both influenza A and SARS-CoV-2 are present in the host remains unclear. Using a mammalian model, sequential infection was performed in ferrets and in K18-hACE2 mice, with SARS-CoV-2 infection following H1N1. We found that co-infection with H1N1 and SARS-CoV-2 extended the duration of clinical manifestation of COVID-19, and enhanced pulmonary damage, but reduced viral shedding of throat swabs and viral loads in the lungs of ferrets. Moreover, mortality was increased in sequentially infected mice compared with single-infection mice. Compared with single-vaccine inoculation, co-inoculation of PiCoVacc (a SARS-CoV-2 vaccine) and the flu vaccine showed no significant differences in neutralizing antibody titers or virus-specific immune responses. Combined immunization effectively protected K18-hACE2 mice against both H1N1 and SARS-CoV-2 infection. Our findings indicated the development of systematic models of co-infection of H1N1 and SARS-CoV-2, which together notably enhanced pneumonia in ferrets and mice, as well as demonstrated that simultaneous vaccination against H1N1 and SARS-CoV-2 may be an effective prevention strategy for the coming winter.

BACKGROUND To identify the efficacy of a design thinking and health care innovation course in improving medical students' self-awareness regarding design thinking metrics. METHODS The assessment of the design thinking mindset was measured pre- and post-course. The target population included medical students at our institution participating in the Innovation & Design Thinking (IDT) course. A paired t-test was used to compare scores from before and after taking the course with p-value set at

Predictions of certifiably robust classifiers remain constant in a neighborhood of a point, making them resilient to test-time attacks with a guarantee. In this work, we present a previously unrecognized threat to robust machine learning models that highlights the importance of training-data quality in achieving high certified adversarial robustness. Specifically, we propose a novel bilevel optimization-based data poisoning attack that degrades the robustness guarantees of certifiably robust classifiers. Unlike other poisoning attacks that reduce the accuracy of the poisoned models on a small set of target points, our attack reduces the average certified radius (ACR) of an entire target class in the dataset. Moreover, our attack is effective even when the victim trains the models from scratch using state-of-the-art robust training methods such as Gaussian data augmentation\cite{cohen2019certified}, MACER\cite{zhai2020macer}, and SmoothAdv\cite{salman2019provably} that achieve high certified adversarial robustness. To make the attack harder to detect, we use clean-label poisoning points with imperceptible distortions. The effectiveness of the proposed method is evaluated by poisoning MNIST and CIFAR10 datasets and training deep neural networks using previously mentioned training methods and certifying the robustness with randomized smoothing. The ACR of the target class, for models trained on generated poison data, can be reduced by more than 30\%. Moreover, the poisoned data is transferable to models trained with different training methods and models with different architectures., CVPR 2021

Breast cancer has the highest incidence among women in Taiwan, which is mainly screened by mammography. When the doctor observes the mammogram and initially judges that the patient has malignant microcalcification (MC) clusters, the patient must undergo needle localization surgical biopsy. However, needle localization surgical biopsy makes the patient painful, and the color of breast tissue and MCs are all white, which makes it difficult for doctors to judge where MCs are clustered immediately. Thus, we use VGG16 to find out breast MC clusters from the image. Moreover, we use Mask RCNN to find MCs from the clusters to remove the noise from the background. Finally, we use Inception V3 to identify the benign and malignant of MC clusters. The accuracy of the cluster classification, MCs labeling and benign and malignant analysis are 93%, 95% and 91%. Furthermore, the precision, specificity and sensitivity of our proposed methods are about 87%, 89% and 90%, respectively. It proved that our system can effectively assist doctors in diagnosing and reduce the burden on patients and medical personnel.

IntroductionAs a structural proxy for evaluating brain health, neuroimaging-based brain age gap (BAG) is presumed to link the dementia risks to cognitive changes in the premorbid phase, but this remains unclear.MethodsBrain age prediction models were constructed and applied to a population-based cohort (N=371) to estimate their BAG. Further, structural equation modeling was employed to investigate the mediation effect of BAG between risk levels (assessed by 2 dementia-related risk scores) and cognitive changes (examined by 4 cognitive assessments).ResultsA higher burden of modifiable dementia risk factors was causally associated with a greater cognitive decline, and this was significantly mediated (P=0.017) by a larger multimodal BAG, which indicated an older brain. Moreover, a steeper slope (P=0.020) of association between cognitive decline and multimodal BAG was observed when individuals had higher dementia risks.DiscussionMultimodal BAG is a potential mediating indicator to reflect the changes in the pathophysiological mechanism of cognitive aging.

Background: Accelerated cell apoptosis is a crucial pathogenic mechanism in lupus nephritis (LN) with dysregulated expression levels of long noncoding RNAs (lncRNAs). The expression of pro-apoptotic lincRNA-p21 and its competing endogenous RNA target miR-181a were studied in LN patients, human kidney cell and T-lymphocyte lines with CRISPR interference-conducted repression and lentiviral vector-mediated overexpression of lincRNA-p21, and a mouse LN model. Methods: Clinical samples were collected from LN patients with higher disease activity and control subjects including lupus patients without renal involvement and age/sex-matched healthy controls (HCs). The expression of lincRNA-p21, H19 (anti-apoptotic lncRNA) and miR-181a were examined in peripheral blood mononuclear cells (PBMNCs) and urine cells, and analyzed for clinical correlation. Cell lines were treated with doxorubicin (Dox) to induce apoptosis and evaluate for the expression of lincRNA-p21, caspase 3 and p21. LincRNA-p21-silened HEK 293T and Jurkat transfectants were examined for apoptosis and miR-181a expression. LincRNA-p21-overexpressed HK-2 cells were examined for apoptosis and p53-related down-stream molecules levels. Female Balb/C mice were injected with pristane to induce LN, and examined for the expression of anti-DNA, proteinuria, lincRNA-p21, caspase 3 and p21 as well as in situ apoptosis. Results: Up-regulated expression of lincRNA-p21 rather than H19 were identified in PBMNCs from LN patients, positively correlated with disease activity and proteinuria amount. Higher lincRNA-p21 levels were identified in LN CD4+T cells than other subpopulations. LN urine cells had greater lincRNA-p21 levels than HCs. There were lower miR-181a levels in PBMNCs from LN patients, negatively correlated with disease activity. Dox-induced apoptotic cell lines had up-regulated levels of lincRNA-p21, caspase 3 and p21, whereas down-regulated miR-181a expression with decreased TCRζchain and IL-2 levels was identified in Jurkat cells. LincRNA-p21-silenced transfectants displayed reduced apoptosis with up-regulated miR-181a expression. LincRNA-p21-overexpressed HK-2 cells revealed enhanced apoptosis with up-regulated expression of downstream PUMA and Bax molecules. LN mice had in situ apoptosis and progressively increased anti-dsDNA, proteinuria and renal lincRNA-p21 levels with up-regulated expression of caspase 3 and p21.Conclusions: By using clinical samples, human cell lines and a mouse model, we demonstrate up-regulated expression of lincRNA-p21 in LN, implicating a potential activity biomarker and therapeutic target.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is highly transmitted through the respiratory route, but potential extra-respiratory routes of SARS-CoV-2 transmission remain uncertain. Here we inoculated five rhesus macaques with 1 × 106 TCID50 of SARS-CoV-2 conjunctivally (CJ), intratracheally (IT), and intragastrically (IG). Nasal and throat swabs collected from CJ and IT had detectable viral RNA at 1–7 days post-inoculation (dpi). Viral RNA was detected in anal swabs from only the IT group at 1–7 dpi. Viral RNA was undetectable in tested swabs and tissues after intragastric inoculation. The CJ infected animal had a higher viral load in the nasolacrimal system than the IT infected animal but also showed mild interstitial pneumonia, suggesting distinct virus distributions. This study shows that infection via the conjunctival route is possible in non-human primates; further studies are necessary to compare the relative risk and pathogenesis of infection through these different routes in more detail., SARS-CoV-2 mainly transmits via respiratory droplets. Here Deng et al. show that SARS-CoV-2 can infect rhesus macaques via ocular conjunctival inoculation.

To achieve a highly reliable DC distribution microgrid network, a bidirectional isolated ripple cancel triple active bridge (TAB) converter is proposed in this paper. In a conventional full-bridge TAB converter, the DC-link capacitors suffer from high ripple current, which significantly reduces their lifetime. To solve this issue, the proposed converter can reduce the ripple current to nearly zero by adding the clamping capacitors as an internal ripple-cancellation circuit topology. In addition, the proposed converter inherits the advantages of the conventional full-bridge TAB converter such as a wide range of soft switching and bidirectional power conversion. This makes it easier to replace the conventional full-bridge TAB converters. Finally, a 1 kW prototype was built to demonstrate its feasibility. In the experiment, the ripple current reduced to nearly zero (0.16 A) under 400 V/400 V/400 V and rated power 1 kW operating conditions. The proposed converter could be used to make highly reliable and efficient DC distribution microgrid as a power router.

ObjectivesTo explore how to integrate the "best" practice into nursing of venous thromboembolism (VTE) based on the integrated-Promoting Action on Research Implementation in Health Services (i-PARIHS) framework.MethodsA mixed-methods design was used. A steering group for clinical evidence implementation (EI) was established to conduct pre-implementation baseline surveys, a thorough analysis of the evidence, and an analysis of the survey results. The hindering and enabling factors associated with the clinical implementation of the evidence were analysed based on the three core elements of i-PARIHS, to formulate the clinical implementation plan for VTE nursing evidence. On-site expert reviews and focus group interviews were used to evaluate the feasibility of the draft plan, make adjustments, and finalize the evidence-based practice plan, which was then put into practice and evaluated.ResultsA new nursing process, a health education manual and a nursing quality checklist on VTE has been established and proved to be appropriate through the implementation. Compliance with evidence related to VTE nursing increased significantly in the two units, with better compliance in unit B than unit A. The knowledge, attitude and behaviour scores for VTE nursing increased substantially in both nurses and patients.ConclusionThe EI programme of incorporating the "best" evidence on VTE nursing into clinical practice using the i-PARIHS framework demonstrated feasibility, appropriateness and effectiveness and could serve as a reference.

Creativity, a process that generates novel and meaningful ideas, involves increased association between task-positive (control) and task-negative (default) networks in the human brain. Inspired by this seminal finding, in this study we propose a creative decoder within a deep generative framework, which involves direct modulation of the neuronal activation pattern after sampling from the learned latent space. The proposed approach is fully unsupervised and can be used off-the-shelf. Several novelty metrics and human evaluation were used to evaluate the creative capacity of the deep decoder. Our experiments on different image datasets (MNIST, FMNIST, MNIST+FMNIST, WikiArt and CelebA) reveal that atypical co-activation of highly activated and weakly activated neurons in a deep decoder promotes generation of novel and meaningful artifacts., Accepted to IJCAI 2020

Verifying robustness of neural networks given a specified threat model is a fundamental yet challenging task. While current verification methods mainly focus on the $\ell_p$-norm threat model of the input instances, robustness verification against semantic adversarial attacks inducing large $\ell_p$-norm perturbations, such as color shifting and lighting adjustment, are beyond their capacity. To bridge this gap, we propose \textit{Semantify-NN}, a model-agnostic and generic robustness verification approach against semantic perturbations for neural networks. By simply inserting our proposed \textit{semantic perturbation layers} (SP-layers) to the input layer of any given model, \textit{Semantify-NN} is model-agnostic, and any $\ell_p$-norm based verification tools can be used to verify the model robustness against semantic perturbations. We illustrate the principles of designing the SP-layers and provide examples including semantic perturbations to image classification in the space of hue, saturation, lightness, brightness, contrast and rotation, respectively. In addition, an efficient refinement technique is proposed to further significantly improve the semantic certificate. Experiments on various network architectures and different datasets demonstrate the superior verification performance of \textit{Semantify-NN} over $\ell_p$-norm-based verification frameworks that naively convert semantic perturbation to $\ell_p$-norm. The results show that \textit{Semantify-NN} can support robustness verification against a wide range of semantic perturbations.

We simulated 3 transmission modes, including close-contact, respiratory droplets and aerosol routes, in the laboratory. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) can be highly transmitted among naive human angiotensin-converting enzyme 2 (hACE2) mice via close contact because 7 of 13 naive hACE2 mice were SARS-CoV-2 antibody seropositive 14 days after being introduced into the same cage with 3 infected-hACE2 mice. For respiratory droplets, SARS-CoV-2 antibodies from 3 of 10 naive hACE2 mice showed seropositivity 14 days after introduction into the same cage with 3 infected-hACE2 mice, separated by grids. In addition, hACE2 mice cannot be experimentally infected via aerosol inoculation until continued up to 25 minutes with high viral concentrations., Severe acute respiratory syndrome coronavirus 2 can be highly transmitted among naive angiotensin-converting enzyme 2 mice via close-contact spreading by respiratory droplets. However, these mice cannot be experimentally infected through aerosol inoculation unless it is continued for 25 minutes with a high viral concentration.

Graph Neural Networks (GNNs) have made significant advances on several fundamental inference tasks. As a result, there is a surge of interest in using these models for making potentially important decisions in high-regret applications. However, despite GNNs' impressive performance, it has been observed that carefully crafted perturbations on graph structures (or nodes attributes) lead them to make wrong predictions. Presence of these adversarial examples raises serious security concerns. Most of the existing robust GNN design/training methods are only applicable to white-box settings where model parameters are known and gradient based methods can be used by performing convex relaxation of the discrete graph domain. More importantly, these methods are not efficient and scalable which make them infeasible in time sensitive tasks and massive graph datasets. To overcome these limitations, we propose a general framework which leverages the greedy search algorithms and zeroth-order methods to obtain robust GNNs in a generic and an efficient manner. On several applications, we show that the proposed techniques are significantly less computationally expensive and, in some cases, more robust than the state-of-the-art methods making them suitable to large-scale problems which were out of the reach of traditional robust training methods., Comment: Accepted by ICASSP 2020

Background: Since December 2019, an outbreak of the Corona Virus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus (SARS-CoV-2) in Wuhan, China, has become a public health emergency of international concern. The high fatality of aged cases caused by SARS-CoV-2 was a need to explore the possible age-related phenomena with non-human primate models. Methods: Three 3-5 years old and two 15 years old rhesus macaques were intratracheally infected with SARS-CoV-2, and then analyzed by clinical signs, viral replication, chest X-ray, histopathological changes and immune response. Results: Viral replication of nasopharyngeal swabs, anal swabs and lung in old monkeys was more active than that in young monkeys for 14 days after SARS-CoV-2 challenge. Monkeys developed typical interstitial pneumonia characterized by thickened alveolar septum accompanied with inflammation and edema, notably, old monkeys exhibited diffuse severe interstitial pneumonia. Viral antigens were detected mainly in alveolar epithelial cells and macrophages. Conclusion: SARS-CoV-2 caused more severe interstitial pneumonia in old monkeys than that in young monkeys. Rhesus macaque models infected with SARS-CoV-2 provided insight into the pathogenic mechanism and facilitated the development of vaccines and therapeutics against SARS-CoV-2 infection. Key words: pathogenicity; pneumonia; rhesus macaque model; SARS-CoV-2

The outbreak of Corona Virus Disease 2019 caused by the severe acute respiratory syndrome coronavirus (SARS-CoV-2) is highly transmitted. The potential extra-respiratory transmission routes remain uncertain. Five rhesus macaques were inoculated with 1×106 TCID50 of SARS-CoV-2 via conjunctival (CJ), intratracheal (IT), and intragastric (IG) routes, respectively. Remarkably, the CJ inoculated-macaques developed mild interstitial pneumonia and viral load was detectable in the conjunctival swabs at 1 days post-inoculation (dpi). Only via IT inoculation, viral load was detected in the anal swab at 1-7 dpi and macaque showed weight loss. However, viral load was undetectable after IG inoculation. Comparatively, viral load was higher in the nasolacrimal system but lesions of lung were relatively mild and local via CJ inoculation compared with that via IT inoculation, demonstrating distinct characteristics of virus dispersion. Both the two routes affected the alimentary tract. Therefore the clinicians need to protect eye while working with patients.

A global pandemic of Corona Virus Disease 2019 (COVID-19) caused by severe acute respiratory syndrome CoV-2 (SARS-CoV-2) is ongoing spread. It remains unclear whether the convalescing patients have a risk of reinfection. Rhesus macaques were rechallenged with SARS-CoV-2 during an early recovery phase from initial infection characterized by weight loss, interstitial pneumonia and systemic viral dissemination mainly in respiratory and gastrointestinal tracts. The monkeys rechallenged with the identical SARS-CoV-2 strain have failed to produce detectable viral dissemination, clinical manifestations and histopathological changes. A notably enhanced neutralizing antibody response might contribute the protection of rhesus macaques from the reinfection by SARS-CoV-2. Our results indicated that primary SARS-CoV-2 infection protects from subsequent reinfection.One Sentence SummaryNeutralizing antibodies against SARS-CoV-2 might protect rhesus macaques which have undergone an initial infection from reinfection during early recovery days.

The prevalence of nontuberculous mycobacterial lung disease (NTM-LD) has increased in Western and Asian nations in recent decades. While surgery may improve the outcome of more complex cases, many inconsistencies exist in the current literature regarding the management, growing emergence, and challenges of drug-resistant forms of NTM-LD, the indications and timing of surgical treatment, and perioperative multimodal therapy of NTM-LD. Moreover, data regarding the comparative treatments, risk factors of pulmonary resection for NTM-LD, and the long-term outcomes of microbiological recurrence are limited. This review will focus on outlining the outcomes of recently optimized surgical approaches, as well as providing an overview of the roles of perioperative multimodalities therapies in the treatment of NTM-LD.

Medium-/long-distance power transmission in future DC power distribution network systems requires medium voltage (1 kV~20 kV) to reduce the transmission losses. Power semiconductor devices of the conventional full-bridge configuration, however, face high voltage stress issues. Several topologies were investigated to solve this issue, including multilevel converters and solid-state transformers. In this study, a bidirectional isolated ripple cancel dual active bridge modular multilevel DC-DC converter was investigated for medium-voltage DC applications, such as the 6.6 k V DC power transmission between two local 800 V DC distribution power network systems. The proposed modular multilevel circuit was developed based on prior research on a bidirectional ripple cancel dual active bridge converter. The proposed converter inherits the latter’s features, including ripple cancel characteristic on DC-link capacitors, soft switching (ZVS), transformer DC bias current elimination, and automatic voltage balancing on series-connected semiconductors. Thus, the modular multilevel DC-DC converter can be expected to have high reliability. This paper discusses the circuit operation, ripple cancel, DC bias current elimination, and voltage balancing features of the proposed circuit. The prototype of the proposed circuit was fabricated at a scaled down version, 900 V/ 600 V/ 900 W, and its basic operation was tested to demonstrate its feasibility.

Recent deep neural networks based techniques, especially those equipped with the ability of self-adaptation in the system level such as deep reinforcement learning (DRL), are shown to possess many advantages of optimizing robot learning systems (e.g., autonomous navigation and continuous robot arm control.) However, the learning-based systems and the associated models may be threatened by the risks of intentionally adaptive (e.g., noisy sensor confusion) and adversarial perturbations from real-world scenarios. In this paper, we introduce timing-based adversarial strategies against a DRL-based navigation system by jamming in physical noise patterns on the selected time frames. To study the vulnerability of learning-based navigation systems, we propose two adversarial agent models: one refers to online learning; another one is based on evolutionary learning. Besides, three open-source robot learning and navigation control environments are employed to study the vulnerability under adversarial timing attacks. Our experimental results show that the adversarial timing attacks can lead to a significant performance drop, and also suggest the necessity of enhancing the robustness of robot learning systems., Accepted to IEEE ICASSP 2020

Despite the great achievements of the modern deep neural networks (DNNs), the vulnerability/robustness of state-of-the-art DNNs raises security concerns in many application domains requiring high reliability. Various adversarial attacks are proposed to sabotage the learning performance of DNN models. Among those, the black-box adversarial attack methods have received special attentions owing to their practicality and simplicity. Black-box attacks usually prefer less queries in order to maintain stealthy and low costs. However, most of the current black-box attack methods adopt the first-order gradient descent method, which may come with certain deficiencies such as relatively slow convergence and high sensitivity to hyper-parameter settings. In this paper, we propose a zeroth-order natural gradient descent (ZO-NGD) method to design the adversarial attacks, which incorporates the zeroth-order gradient estimation technique catering to the black-box attack scenario and the second-order natural gradient descent to achieve higher query efficiency. The empirical evaluations on image classification datasets demonstrate that ZO-NGD can obtain significantly lower model query complexities compared with state-of-the-art attack methods., accepted by AAAI 2020

This tutorial will teach participants to use and contribute to a new open-source Python package named AI Explainability 360 (AIX360) (https://aix360.mybluemix.net), a comprehensive and extensible toolkit that supports interpretability and explainability of data and machine learning models. Motivation for the toolkit. The AIX360 toolkit illustrates that there is no single approach to explainability that works best for all situations. There are many ways to explain: data vs. model, direct vs. post-hoc explanation, local vs. global, etc. The toolkit includes ten state of the art algorithms that cover different dimensions of explanations along with proxy explainability metrics. Moreover, one of our prime objectives is for AIX360 to serve as an educational tool even for non-machine learning experts (viz. social scientists, healthcare experts). To this end, the toolkit has an interactive demonstration, highly descriptive Jupyter notebooks covering diverse real-world use cases, and guidance materials, all helping one navigate the complex explainability space. Compared to existing open-source efforts on AI explainability, AIX360 takes a step forward in focusing on a greater diversity of ways of explaining, usability in industry, and software engineering. By integrating these three aspects, we hope that AIX360 will attract researchers in AI explainability and help translate our collective research results for practicing data scientists and developers deploying solutions in a variety of industries. Regarding the first aspect of diversity, Table 1 in [1] compares AIX360 to existing toolkits in terms of the types of explainability methods offered. The table shows that AIX360 not only covers more types of methods but also has metrics which can act as proxies for judging the quality of explanations. Regarding the second aspect of industry usage, AIX360 illustrates how these explainability algorithms can be applied in specific contexts (please see Audience, goals, and outcomes below). In just a few months since its initial release, the AIX360 toolkit already has a vibrant slack community with over 120 members and has been forked almost 80 times accumulating over 400 stars. This response leads us to believe that there is significant interest in the community in learning more about the toolkit and explainability in general. Audience, goals, and outcomes. The presentations in the tutorial will be aimed at an audience with different backgrounds and computer science expertise levels. For all audience members and especially those unfamiliar with Python programming, AIX360 provides an interactive experience (http://aix360.mybluemix.net/data) centered around a credit approval scenario as a gentle and grounded introduction to the concepts and capabilities of the toolkit. We will also teach all participants which type of explainability algorithm is most appropriate for a given use case, not only for those in the toolkit but also from the broader explainability literature. Knowing which explainability algorithms apply to which contexts and understanding when to use them can benefit most people, regardless of their technical background. The second part of the tutorial will consist of three use cases featuring different industry domains and explanation methods. Data scientists and developers can gain hands-on experience with the toolkit by running and modifying Jupyter notebooks, while others will be able to follow along by viewing rendered versions of the notebooks. Here is a rough agenda of the tutorial: 1) Overture: Provide a brief introduction to the area of explainability as well as introduce common terms. 2) Interactive Web Experience: The AIX360 interactive web experience (http://aix360.mybluemix.net/data) is intended to show a non-computer science audience how different explainability methods may suit different stakeholders in a credit approval scenario (data scientists, loan officers, and bank customers). 3) Taxonomy: We will next present a taxonomy that we have created for organizing the space of explanations and guiding practitioners toward an appropriate choice for their applications. 4) Installation: We will transition into a Python environment and ask participants to install the AIX360 package on their machines using provided instructions. 5) Example Use Cases in Finance, Government, and Healthcare: We will take participants through three use-cases in various application domains in the form of Jupyter notebooks. 6) Metrics: We will briefly showcase the two explainability metrics currently available through the toolkit. 7) Future Directions: The final segment will be to discuss future directions and how participants can contribute to the toolkit.

With the rapid development of deep learning techniques in computer vision, emotion detection has been an emerging topic for different industries, including healthcare, mobile entertainment, and even securities. These organizations leverage emotion detection and apply it to both real-time cameras and offline videos to provide more interactivity between users and devices. In this study, we provide a lightweight personalized accompanying system for home security using emotion detection and IoT devices. We implement our design and demonstrate a practical scenario using Nvidia Jetson Nano. According to our design, we hope the system can benefit the health monitoring and interactivity for those elderly living alone.