A Hybrid Model of Deep Learning Features and Clinical Features for Severe Cases Predication of COVID-19

COVID-19 has spread around the world since 2019. Approximately 6.5% of COVID-19 a risk of developing severe disease with high mortality rate. To reduce the mortality rate and provide appropriate treatment, this research established an integrated models with to predict the clinical outcome of COVID-19 patients with clinical, deep learning and radiomics features. To obtain the optimal feature combination for prediction, 9 clinical features combination was selected from all available clinical factors after using LASSO, 18 deep learning features from U-Net architecture, and 9 radiomics features from segmentation result. A total of 213 COVID-19 patients and 335 non-COVID-19 patients from 5 hospitals were enrolled and used as training and test sample in this research. The proposed model obtained an accuracy, precision, recall, specificity, F1-score and ROC curve of 0.971, 0.943, 0.937, 0.974, 0.941 and 0.979, respectively, which exceeds the related work using only clinical, deep learning or radiomics factors. © 2023 SPIE.

Hierarchical Multi-agent Model for Reinforced Medical Resource Allocation with Imperfect Information

With the advent of the COVID-19 pandemic, the shortage in medical resources became increasingly more evident. Therefore, efficient strategies for medical resource allocation are urgently needed. However, conventional rule-based methods employed by public health experts have limited capability in dealing with the complex and dynamic pandemic-spreading situation. In addition, model-based optimization methods such as dynamic programming (DP) fail to work since we cannot obtain a precise model in real-world situations most of the time. Model-free reinforcement learning (RL) is a powerful tool for decision-making;however, three key challenges exist in solving this problem via RL: (1) complex situations and countless choices for decision-making in the real world;(2) imperfect information due to the latency of pandemic spreading;and (3) limitations on conducting experiments in the real world since we cannot set up pandemic outbreaks arbitrarily. In this article, we propose a hierarchical RL framework with several specially designed components. We design a decomposed action space with a corresponding training algorithm to deal with the countless choices, ensuring efficient and real-time strategies. We design a recurrent neural network-based framework to utilize the imperfect information obtained from the environment. We also design a multi-agent voting method, which modifies the decision-making process considering the randomness during model training and, thus, improves the performance. We build a pandemic-spreading simulator based on real-world data, serving as the experimental platform. We then conduct extensive experiments. The results show that our method outperforms all baselines, which reduces infections and deaths by 14.25% on average without the multi-agent voting method and up to 15.44% with it.

The effect of Pickering emulsion adjuvants on the immune efficacy of the COVID-19 polypeptide vaccine

Pickering emulsions were prepared by phacoemulsification in an ice water bath with squalene as the oil phase and an aluminum adjuvant as the particle stabilizer. The effects of formulation and process conditions on the size and distribution of the Pickering emulsions were investigated. Pickering emulsions prepared under the optimal prescription and process conditions were mixed with a peptide antigen to obtain a peptide vaccine. The optimal prescription and process condition of the Pickering emulsion is as follows: squalene as the oil phase, ultra-pure water as the water phase with 5 mg/ml aluminum adjuvant, and an ultrasonication time of 4 min at 200 W power. BALB/c mice were immunized with the peptide vaccine, and the ability of the Pickering emulsion as an immunological adjuvant to improve the efficacy of the peptide vaccine was evaluated. Under optimal conditions, a Pickering emulsion with a small particle size (430.8 nm), uniform distribution (polydispersion index of 16.9%), and zeta potential of 31.5 mV, was obtained. Immunological results showed that the serum specific antibody level in the vaccinated group reached 1×104 after three immunizations. The proportion of CD4+T cells and CD4/CD8 cells was significantly higher (P < 0.05) in the vaccinated groups than the blank control group. Further, cytokine (TNF-α) secretion decreased in the aluminum adjuvant and Pickering emulsion groups but increased in the Freund's adjuvant group. All three vaccinated groups of mice exhibited low but detectable levels of IFN-γ secretion. © 2023

Hierarchical Multi-agent Model for Reinforced Medical Resource Allocation with Imperfect Information

With the advent of the COVID-19 pandemic, the shortage in medical resources became increasingly more evident. Therefore, efficient strategies for medical resource allocation are urgently needed. However, conventional rule-based methods employed by public health experts have limited capability in dealing with the complex and dynamic pandemic-spreading situation. In addition, model-based optimization methods such as dynamic programming (DP) fail to work since we cannot obtain a precise model in real-world situations most of the time. Model-free reinforcement learning (RL) is a powerful tool for decision-making;however, three key challenges exist in solving this problem via RL: (1) complex situations and countless choices for decision-making in the real world;(2) imperfect information due to the latency of pandemic spreading;and (3) limitations on conducting experiments in the real world since we cannot set up pandemic outbreaks arbitrarily. In this article, we propose a hierarchical RL framework with several specially designed components. We design a decomposed action space with a corresponding training algorithm to deal with the countless choices, ensuring efficient and real-time strategies. We design a recurrent neural network-based framework to utilize the imperfect information obtained from the environment. We also design a multi-agent voting method, which modifies the decision-making process considering the randomness during model training and, thus, improves the performance. We build a pandemic-spreading simulator based on real-world data, serving as the experimental platform. We then conduct extensive experiments. The results show that our method outperforms all baselines, which reduces infections and deaths by 14.25% on average without the multi-agent voting method and up to 15.44% with it. © 2022 Association for Computing Machinery.

Survey of Telecommuting System Security

Affected by the Corona Virus Disease 2019 (COVID-19), telecommuting, a new type of office, has developed rapidly in a short period of time and has been widely used in society, and the resulting security problems of telecommuting systems have become more and more urgent and prominent. At present, the research on the security of telecommuting systems is still in its infancy, and the research results are not enough to completely solve the security problems in the development of telecommuting systems. In order to systematically understand the current research progress researchers, this paper summarizes the security problems of telecommuting systems for the first time, and writes this review. This paper first reviews the development process of the telecommuting system, points out the unique security requirements and problems of the telecommuting system in different application scenarios, and then divides the telecommuting system into virtual private network (VPN), remote desktop control and teamwork platform, according to the technical architecture of the telecommuting system. Based on nearly 5 years of research on telecommuting papers published in the EI Database, Web of Science database and CCF recommended international conference on network and information security, as well as other related high-level research work, this paper systematically analyzes and summarizes the security problems existing in the above three types of telecommuting systems, especially focusing on the security problems of teamwork platforms, a new type of telecommuting. According to the architecture and function of the teamwork platform and the attack methods commonly used by attackers, the security risk of teamwork platforms are divided into five categories: third-party APP security, communication protocol security, client security, cloud server security, and side channel analysis. Finally, the challenges and opportunities faced by the telecommuting system security research institute are pointed out, and the direction for the future research of telecommuting system security is pointed out. © 2022 Chinese Academy of Sciences. All rights reserved.

Optimization Model and Algorithm of Logistics Vehicle Routing Problem under Major Emergency

The novel coronavirus pandemic is a major global public health emergency, and has presented new challenges and requirements for the timely response and operational stability of emergency logistics that were required to address the major public health events outbreak in China. Based on the problems of insufficient timeliness and high total system cost of emergency logistics distribution in major epidemic situations, this paper takes the minimum vehicle distribution travel cost, time cost, early/late punishment cost, and fixed cost of the vehicle as the target, the soft time window for receiving goods at each demand point, the rated load of the vehicle, the volume, maximum travel of the vehicle in a single delivery as constraints, and an emergency logistics vehicle routing problem optimization model for major epidemics was constructed. The convergence speed improvement strategy, particle search improvement strategy, and elite retention improvement strategy were introduced to improve the particle swarm optimization (PSO) algorithm for it to be suitable for solving global optimization problems. The simulation results prove that the improved PSO algorithm required to solve the emergency medical supplies logistics vehicle routing problem for the major emergency can reach optimal results. Compared with the basic PSO algorithm, the total cost was reduced by 20.09%. © 2023 by the authors.

[Mechanism and clinical treatment strategy of heparin resistance].

Heparin resistance is becoming a hot issue of clinical concern. In critically ill patients, heparin resistance can lead to failure of anticoagulation therapy or increase the risk of major bleeding. Prompt recognition of heparin resistance can help to precisely adjust heparin dosage and avoid deterioration and adverse events. Heparin resistance can be mechanistically classified into the antithrombin-mediated and the non-antithrombin-mediated. Common etiologies include heparin-induced thrombocytopenia, severe infections such as severe COVID-19, treatment with extracorporeal circulation or extracorporeal membrane oxygenation (ECMO), and use of factor Xa reversal agents; heparin resistance is now often identified by the concordance of activated partial thromboplastin time (APTT) ratio with anti-FXa. Common clinical management strategies include antithrombin supplementation and replacement of anticoagulant drugs (e.g., direct thrombin inhibitors), but their safety and efficacy still need to be further validated.

Backward bifurcation, basic reinfection number and robustness of an SEIRE epidemic model with reinfection

Recent evidences show that individuals who recovered from COVID-19 can be reinfected. However, this phenomenon has rarely been studied using mathematical models. In this paper, we propose an SEIRE epidemic model to describe the spread of the epidemic with reinfection. We obtain the important thresholds R-0 (the basic reproduction number) and R-c (a threshold less than one). Our investigations show that when R-0 > 1, the system has an endemic equilibrium, which is globally asymptotically stable. When R-c < R-0 < 1, the epidemic system exhibits bistable dynamics. That is, the system has backward bifurcation and the disease cannot be eradicated. In order to eradicate the disease, we must ensure that the basic reproduction number R0 is less than Rc. The basic reinfection number is obtained to measure the reinfection force, which turns out to be a new tipping point for disease dynamics. We also give definition of robustness, a new concept to measure the difficulty of completely eliminating the disease for a bistable epidemic system. Numerical simulations are carried out to verify the conclusions.

Mortgage Appraisal Waivers and Prepayment Speeds

This report examines factors affecting the use of appraisal waivers for mortgages guaranteed by Fannie Mae and Freddie Mac and the effect of appraisal waivers on prepayment speeds. It shows that the alignment of Freddie Mac's eligibility criteria with those of Fannie Mae around the start of the COVID-19 pandemic was associated with an increase in the use of appraisal waivers. Conditional on satisfying the basic eligibility criteria, appraisal waivers are more common for refinance loans, loans serviced by nonbanks, and less risky borrowers. The report also shows that appraisal waivers were associated with higher conditional prepayment rates during 2020 but to a lesser extent in 2021 as refinancing activity slowed down. Much of this association can be explained by correlations between appraisal waivers and other observable determinants of prepayment speeds.

Molnupiravir-an oral drug for COVID-19 and its comparative analysis

Objective: The mechanism of action, metabolic kinetics, efficacy, safety and drug-drug interaction of molnupiravir were reviewed to provide a basis for clinical use. Method(s): Literature related to molnupiravir was systematically searched in Chinese Clinical Trial Registry, clinicaltrials.gov, Pubmed, Chinese Journal Full-text Database (CNKI) and Wanfang database, and the relevant information was reviewed. Results & Conclusion(s): Molnupiravir was the world's first small-molecule oral drug for COVID-19, which had been approved or authorized for emergency use in more than 40 countries all over the world. Molnupiravir was a ribonucleoside analogue that could be caused mutations in RNA products by viral RNA polymerase, and thus halt viral replication. Clinical trial results showed that molnupiravir could be reduced hospitalization and mortality rates in patients with mild and moderate COVID-19, and might be effective against SARS-CoV-2 mutant strains.Molnupiravir had good safety and tolerability, to provide reference for the treatment of COVID-19 in the future. Copyright © 2022, Chinese Journal of New Drugs Co. Ltd. All right reserved.

Research progress of mRNA vaccine delivery system

Messenger ribonucleic acid (mRNA) vaccine is characteristic of subunit vaccines and live attenuated vectors. It can simultaneously induce humoral and cellular immunity, and has the advantage of easy to produce on a large scale. Therefore, great application prospects are seen in the field of infectious diseases and the treatment of major diseases. In the field of biomedicine, mRNA vaccine has been deeply cultivated for decades. With the outbreak of the COVID-19 epidemic and the pandemic, the process of listing mRNA vaccines has been greatly accelerated and promoted. The COVID-19 mRNA vaccines, with lipid nanoparticles (LNPs) as the carrier which deliver the mRNA encoding the COVID-19 spike protein antigen information to host cells, were successfully launched, showing good safety and efficacy. Meanwhile, the development of the delivery system, as the key technology of mRNA vaccines, has also become a "battlefield" for the development of mRNA technology. Various delivery technologies, including LNPs, are being actively developed in China and abroad. Safe and effective delivery with independent intellectual property rights will become the key to promote the marketing of mRNA products, which will obtain benefits in the future. This article reviews the mRNA vaccine delivery system and research progress. Copyright © 2022, Chinese Journal of New Drugs Co. Ltd. All right reserved.

The optimal anticoagulation strategy for COVID-19, prophylactic or therapeutic?: a meta-analysis, trial sequential analysis, and meta-regression of more than 27,000 participants

Background: Anticoagulants are promising regimens for treating coronavirus disease 2019 (COVID-19). However, whether prophylactic or intermediate-to-therapeutic dosage is optimal remains under active discussion. Methods: We comprehensively searched PubMed, Embase, Scopus, Web of Science, Cochrane Library, ClinicalTrials, and MedRxiv databases on April 26, 2022. Two independent researchers conducted literature selection and data extraction separately according to predetermined criteria. Notably, this is the first meta-analysis on COVID-19, taking serious consideration regarding the dosage overlap between the 2 comparison groups of prophylactic anticoagulation (PA) and intermediate-to-therapeutic anticoagulation (I-TA). Results: We included 11 randomized controlled trials (RCTs) and 36 cohort studies with 27,051 COVID-19 patients. By analyzing all the RCTs, there was no significant difference in mortality between the PA and I-TA groups, which was further confirmed by trial sequential analysis (TSA) (odds ratio [OR]: 0.93;95% confidence interval [CI]: 0.71–1.22;P = 0.61;TSA adjusted CI: 0.71–1.26). The rate of major bleeding was remarkably higher in the I-TA group than in the PA group, despite adjusting for TSA (OR: 1.73;95% CI: 1.15–2.60;P = 0.009;TSA adjusted CI: 1.09–2.58). RCTs have supported the beneficial effect of I-TA in reducing thrombotic events. After including all studies, mortality in the I-TA group was significantly higher than in the PA group (OR: 1.38;95% CI: 1.15–1.66;P = 0.0005). The rate of major bleeding was similar to the analysis from RCTs (OR: 2.24;95% CI: 1.86–2.69;P < 0.00001). There was no distinct difference in the rate of thrombotic events between the 2 regimen groups. In addition, in both critical and noncritical subgroups, I-TA failed to reduce mortality but increased major bleeding rate compared with PA, as shown in meta-analysis of all studies, as well as RCTs only. Meta-regression of all studies suggested that there was no relationship between the treatment effect and the overall risk of mortality or major bleeding (P = 0.14, P = 0.09, respectively). Conclusion: I-TA is not superior to PA for treating COVID-19 because it fails to lower the mortality rate but increases the major bleeding rate in both critical and noncritical patients. Copyright © 2022 Shandong University, published by Wolters Kluwer, Inc.

Reinforcement Learning Enhances the Experts: Large-scale COVID-19 Vaccine Allocation with Multi-factor Contact Network

In the fight against the COVID-19 pandemic, vaccines are the most critical resource but are still in short supply around the world. Therefore, efficient vaccine allocation strategies are urgently called for, especially in large-scale metropolis where uneven health risk is manifested in nearby neighborhoods. However, there exist several key challenges in solving this problem: (1) great complexity in the large scale scenario adds to the difficulty in experts' vaccine allocation decision making;(2) heterogeneous information from all aspects in the metropolis' contact network makes information utilization difficult in decision making;(3) when utilizing the strong decision-making ability of reinforcement learning (RL) to solve the problem, poor explainability limits the credibility of the RL strategies. In this paper, we propose a reinforcement learning enhanced experts method. We deal with the great complexity via a specially designed algorithm aggregating blocks in the metropolis into communities and we hierarchically integrate RL among the communities and experts solution within each community. We design a self-supervised contact network representation algorithm to fuse the heterogeneous information for efficient vaccine allocation decision making. We conduct extensive experiments in three metropolis with real-world data and prove that our method outperforms the best baseline, reducing 9.01% infections and 12.27% deaths.We further demonstrate the explainability of the RL model, adding to its credibility and also enlightening the experts in turn. © 2022 Owner/Author.

Early prediction model for disease progression of COV1D-19 patients based on XGBoost: Establishment and evaluation

Objective To construct an XGBoost prediction model to predict disease severity of COVID-19 based on clinical characteristics dataset of COVID-19 patients.Methods A total of 347 laboratory-confirmed COVID-19 patients with complete medical information admitted from Feb 10 to April 5, 2020 were screened from the medical record system of Huoshenshan Hospital.Firstly, 21 features with significant differences were screened out as input features for the training model.Bayesian optimization was performed on the constructed XGBoost model to adjust the parameters, and the optimal combination of features was filtered based on feature importance.To further analyze the positive and negative effects of the numerical size of each feature on the prediction results, each feature importance was quantified and attributed by using SHapley Additive explanations (SHAP).Finally, the performance of the XGBoost prediction model was evaluated, and the model was compared and discussed with other machine learning methods, including support vector machine (SVM), naive Bayes ( NB ) , logical regression ( LR) , and k-nearest neighbors ( KNN ).Results In this study, 21 features with significant differences between the severe and non-severe groups were selected for training and validation.The optimal subset with 10 features in the k-nearest neighbor model obtained the highest value of area under curve ( AUG) among the 4 models in the validation set.XGBoost and support vector machine were better than other machine learning methods in terms of prediction performance (AUG;0.942 0, and 0.959 4 on the test set, respectively) , and the training speed of XGBoost was significantly faster.Conclusion A prediction model based on XGBoost is successfully built to achieve early prediction of disease severity of GOVID-19 patients. © 2022 Journal of Army Medical University. All rights reserved.

PIXEL-LEVEL AND AFFINITY-LEVEL KNOWLEDGE DISTILLATION FOR UNSUPERVISED SEGMENTATION OF COVID-19 LESIONS

Automatic segmentation of COVID-19 lesions is essential for computer-aided diagnosis. However, this task remains challenging because widely-used supervised based methods require large-scale annotated data that is difficult to obtain. Although an unsupervised method based on anomaly detection has shown promising results in [1], its performance is relatively poor. We address this problem by proposing a pixel-level and affinity-level knowledge distillation method. It obtains a pre-trained teacher network with rich semantic knowledge of CT images by constructing and training an auto-encoder at first, and then trains a student network with the same architecture as the teacher by distilling the teacher's knowledge only from normal CT images, and finally localizes COVID-19 lesions using the feature discrepancy between the teacher and the student networks. Besides, except for the traditional pixel-level distillation, we design the affinity-level distillation that takes into account the pairwise relationship of features to fully distill effective knowledge. We evaluate this method by using three different COVID-19 datasets and the experimental results show that the segmentation performance is largely improved when it is compared with the other existing unsupervised anomaly detection methods. © 2022 IEEE

[Annual review of venous thromboembolism in 2021].

In the past year, significant progress has been made in the field of venous thromboembolism (VTE) including risk assessment and anticoagulation prevention, diagnostic strategies and model exploration, new drug development and disease management. Particularly, major breakthroughs have been made in the prevention of VTE with FXI inhibitors and the prevention of novel coronavirus pneumonia with coagulation alterations and anticoagulation interventions. Here, we reviewed the progress and achievements in the field of VTE in the past year, aiming to provide evidence and ideas for the diagnosis, treatment and future studies of VTE.

Smoothing of time-varying graph with the generalized LASSO

This paper proposes a joint model based on the generalized LASSO to smooth a time-varying graph. The model generalizes the gLASSO from a purely spatial setting to a spatial-temporal one. In the proposed model, the first term measures the fitting error, while the second term incorporates the structural information of graphs and total variations of time sequence, and hence the model can extract both temporal and spatial information. To illustrate the performance of the proposed model, we analyzed the simulated datasets for epidemic diseases and the real datasets for COVID-19 and mortality rate in mainland China. The results show that the proposed model can capture the trends/regions simultaneously in both temporal and spatial domains, being an effective model to analyze the problems that can be modelled as time-varying graphs. Author

Materials in advanced design of personal protective equipment: a review.

The outbreak of the Covid-19 pandemic has aroused tremendous attention toward personal protective equipment (PPE) in both scientific research and industrial manufacture. Despite decades of development in PPE design and fabrication, there's still much room for further optimization, in terms, of both protection performance and wear comfort. Interdisciplinary efforts have been devoted to this research field in recent years. Significantly, the innovation of materials, which brings about improved performance and versatile new functions for PPEs, has been widely adopted in PPE design. In this minireview, recent progress in the development of novel materials and structural designs for PPE application are presented in detail with the introduction of various material-based strategies for different PPE types, as well as the examples, which apply auxiliary components into face masks to enrich the functionalities and improve the personal feelings in the pandemic period.