Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 20 de 95
Filter
1.
Lancet Microbe ; 2022.
Article in English | ScienceDirect | ID: covidwho-2031776

ABSTRACT

BACKGROUND: The H3N8 avian influenza virus (AIV) has been circulating in wild birds, with occasional interspecies transmission to mammals. The first human infection of H3N8 subtype occurred in Henan Province, China, in April, 2022. We aimed to investigate clinical, epidemiological, and virological data related to a second case identified soon afterwards in Hunan Province, China. METHODS: We analysed clinical, epidemiological, and virological data for a 5-year-old boy diagnosed with H3N8 AIV infection in May, 2022, during influenza-like illness surveillance in Changsha City, Hunan Province, China. H3N8 virus strains from chicken flocks from January, 2021, to April, 2022, were retrospectively investigated in China. The genomes of the viruses were sequenced for phylogenetic analysis of all the eight gene segments. We evaluated the receptor-binding properties of the H3N8 viruses by using a solid-phase binding assay. We used sequence alignment and homology-modelling methods to study the effect of specific mutations on the human receptor-binding properties. We also conducted serological surveillance to detect the H3N8 infections among poultry workers in the two provinces with H3N8 cases. FINDINGS: The clinical symptoms of the patient were mild, including fever, sore throat, chills, and a runny nose. The patient's fever subsided on the same day of hospitalisation, and these symptoms disappeared 7 days later, presenting mild influenza symptoms, with no pneumonia. An H3N8 virus was isolated from the patient's throat swab specimen. The novel H3N8 virus causing human infection was first detected in a chicken farm in Guangdong Province in December, 2021, and subsequently emerged in several provinces. Sequence analyses revealed the novel H3N8 AIVs originated from multiple reassortment events. The haemagglutinin gene could have originated from H3Ny AIVs of duck origin. The neuraminidase gene belongs to North American lineage, and might have originated in Alaska (USA) and been transferred by migratory birds along the east Asian flyway. The six internal genes had originated from G57 genotype H9N2 AIVs that were endemic in chicken flocks. Reassortment events might have occurred in domestic ducks or chickens in the Pearl River Delta area in southern China. The novel H3N8 viruses possess the ability to bind to both avian-type and human-type sialic acid receptors, which pose a threat to human health. No poultry worker in our study was positive for antibodies against the H3N8 virus. INTERPRETATION: The novel H3N8 virus that caused human infection had originated from chickens, a typical spillover. The virus is a triple reassortment strain with the Eurasian avian H3 gene, North American avian N8 gene, and dynamic internal genes of the H9N2 viruses. The virus already possesses binding ability to human-type receptors, though the risk of the H3N8 virus infection in humans was low, and the cases are rare and sporadic at present. Considering the pandemic potential, comprehensive surveillance of the H3N8 virus in poultry flocks and the environment is imperative, and poultry-to-human transmission should be closely monitored. FUNDING: National Natural Science Foundation of China, National Key Research and Development Program of China, Strategic Priority Research Program of the Chinese Academy of Sciences, Hunan Provincial Innovative Construction Special Fund: Emergency response to COVID-19 outbreak, Scientific Research Fund of Hunan Provincial Health Department, and the Hunan Provincial Health Commission Foundation.

2.
13th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics, BCB 2022 ; 2022.
Article in English | Scopus | ID: covidwho-2029547

ABSTRACT

Fast growing global connectivity and urbanisation increases the risk of spreading worldwide disease. The worldwide SARS-COV-2 disease causes healthcare system strained, especially for the intensive care units. Therefore, prognostic of patients' need for intensive care units is priority at the hospital admission stage for efficient resource allocation. In the early hospitalization, patient chest radiography and clinical data are always collected to diagnose. Hence, we proposed a clinical data structured graph Markov neural network embedding with computed radiography exam features (CGMNN) to predict the intensive care units demand for COVID patients. The study utilized 1,342 patients' chest computed radiography with clinical data from a public dataset. The proposed CGMNN outperforms baseline models with an accuracy of 0.82, a sensitivity of 0.82, a precision of 0.81, and an F1 score of 0.76. © 2022 ACM.

3.
Sci Rep ; 12(1):15496, 2022.
Article in English | PubMed | ID: covidwho-2028728

ABSTRACT

Since late 2019, the outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and the resultant spread of COVID-19 have given rise to a worldwide health crisis that is posing great challenges to public health and clinical treatment, in addition to serving as a formidable threat to the global economy. To obtain an effective tool to prevent and diagnose viral infections, we attempted to obtain human antibody fragments that can effectively neutralize viral infection and be utilized for rapid virus detection. To this end, several human monoclonal antibodies were isolated by bio-panning a phage-displayed human antibody library, Tomlinson I. The selected clones were demonstrated to bind to the S1 domain of the spike glycoprotein of SARS-CoV-2. Moreover, clone A7 in Fab and IgG formats were found to effectively neutralize the binding of S protein to angiotensin-converting enzyme 2 in the low nM range. In addition, this clone was successfully converted to quench-based fluorescent immunosensors (Quenchbodies) that allowed antigen detection within a few minutes, with the help of a handy fluorometer.

4.
Frontiers in Environmental Science ; 10, 2022.
Article in English | Scopus | ID: covidwho-2022691

ABSTRACT

Since the financial crisis, especially after the outbreak of COVID-19, the global trade division of labor has been changing rapidly. The global value chain (GVC) keeps shrinking while the domestic value chain (DVC) continues to develop. Therefore, it is important to re-examine the impact of different modes of value chain division of labor on China’s energy efficiency. In this study, we first constructed an input-output model of provinces embedded in the world to measure the dual embedding of domestic and global value chains. Then we used a three-dimensional fixed-effect model to study the impact and mechanism of dual value chain embedding on energy efficiency. We found that domestic value chain embedding inhibits energy efficiency improvement and global value chain embedding promotes energy efficiency improvement. A series of robustness and endogeneity tests support these findings. The heterogeneity tests revealed that the effects of dual value chain embedding on energy efficiency are more pronounced in low-polluting industries, high-tech industries, years before 2008, and coastal regions. The mechanism test revealed that DVC embedding inhibits energy efficiency by exacerbating the low-end lock-in effect, reducing environmental regulation and scale efficiency, and it increases energy efficiency by increasing technological progress efficiency. GVC embedding improves energy efficiency by weakening the low-end lock-in effect and increasing technical efficiency and scale efficiency, and it inhibits energy efficiency improvement by reducing environmental regulation and technological progress efficiency. Copyright © 2022 Chen, Cheng, Gao and Li.

5.
Frontiers in Bioengineering and Biotechnology ; 10, 2022.
Article in English | Web of Science | ID: covidwho-2022646

ABSTRACT

Digital PCR is the most advanced PCR technology. However, due to the high price of the digital PCR analysis instrument, this powerful nucleic acid detection technology is still difficult to be popularized in the general biochemistry laboratory. Moreover, one of the biggest disadvantages of commercial digital PCR systems is the poor versatility of reagents: each instrument can only be used for a few customized kits. Herein, we built a low-cost digital PCR system. The system only relies on low-cost traditional flat-panel PCR equipment to provide temperature conditions for commercial dPCR chips, and the self-made fluorescence detection system is designed and optically optimized to meet a wide range of reagent requirements. More importantly, our system not only has a low cost (< 8000 US dollars) but also has a much higher universality for nucleic acid detection reagents than the traditional commercial digital PCR system. In this study, several samples were tested. The genes used in the experiment were plasmids containing UPE-1a fragment, TP53 reference DNA, hepatitis B virus DNA, leukemia sample, SARS-COV-2 DNA, and SARS-COV-2 RNA. Under the condition that DNA can be amplified normally, the function of the dPCR system can be realized with simpler and low-price equipment. Some DNA cannot be detected by using the commercial dPCR system because of the special formula when it is configured as the reaction solution, but these DNA fluorescence signals can be clearly detected by our system, and the concentration can be calculated. Our system is more applicable than the commercial dPCR system to form a new dPCR system that is smaller and more widely applicable than commercially available machinery.

6.
PLoS ONE [Electronic Resource] ; 17(8):e0273429, 2022.
Article in English | MEDLINE | ID: covidwho-2021917

ABSTRACT

INTRODUCTION: Public health responses were triggered while COVID-19 was spreading. China redeployed healthcare workers to serve the most vulnerable populations and communities in the initial epicentre-Wuhan. However, it is not known how redeployment processes impacted on healthcare workers in a pandemic crisis. AIMS: To explore the experiences and needs of frontline healthcare workers who were redeployed to care for COVID-19 patients in Wuhan, China, and understand the long-term impacts of the redeployment experience on their work and life. METHODS: A qualitative study was conducted with redeployed healthcare workers using semi-structured interviews and thematic analysis. This study is reported in accordance with the consolidated criteria for reporting qualitative research (COREQ) guidelines. FINDINGS: A total of 20 redeployed healthcare workers (13 nurses and seven physicians) participated, and four themes were generated: (1) Initial feelings and emotions of redeployment-Participants experienced worries and concerns, a sense of isolation and loneliness on their arrival to the epicentre. (2) 'It is like a war zone'-Healthcare workers faced a range of risks and challenges of caring for COVID-19 patients in Wuhan in the context of resource strain. (3) Uncertainty and coping strategies in patient care-Despite the hardships experienced, participants continued to deliver high-quality patient care including psychological care and palliative care, good communication and building mutual trusting relationships. (4) Reflection and far-reaching impacts of caring for COVID-19 patients-Participants felt motivated and encouraged as efforts were recognised by the government and wider society. CONCLUSIONS: Redeployed healthcare workers shared their unique needs and experiences of coping with redeployment and challenges they faced in the context of resource strain, which has significant implications for policy and future practice. The reality of a pandemic may reduce healthcare workers' willingness to work due to various reasons including inadequate preparedness of facilities and workplace safety. It is important to support frontline healthcare workers in order to maintain an adequate healthcare workforce in pandemic crises. Continuously evolving pandemic circumstances and uncertainty highlight the importance of an organized national pandemic response plan for subsequent waves of COVID-19 and future pandemics.

7.
BMC Health Serv Res ; 22(1):1143, 2022.
Article in English | PubMed | ID: covidwho-2021287

ABSTRACT

BACKGROUND: At the end of 2019, the Coronavirus Disease 2019 (COVID-19) pandemic broke out. As front-line health professionals, primary care doctors play a significant role in screening SARS-CoV-2 infection and transferring suspected cases. However, the performance of primary care doctors is influenced by their knowledge and role perception. A web-based cross-sectional survey was conducted to assess the consistency and influencing factors of primary care doctor's role perception and expert advice in the guidelines (regulatory definition). METHODS: We designed the questionnaire using "Wenjuanxing" platform, distributed and collected the questionnaire through WeChat social platform, and surveyed 1758 primary care doctors from 11 community health service stations, community health service centers and primary hospitals in Zhejiang Province, China. After the questionnaire was collected, descriptive statistics were made on the characteristics of participants, and univariate analysis and multivariate analysis were used to determine the relevant factors affecting their role cognition. RESULTS: In the reporting and referral suspected cases and patients receiving treatment, most participants' cognition of their roles were consistent with the requirements of guidelines. However, 49.54% and 61.43% of participant doctors were not in line with the government guidelines for diagnosing and classifying COVID-19 and treating suspected cases, respectively. Having a middle or senior professional title and participating in front-line COVID-19 prevention and control work is beneficial to the accurate role perception of diagnosis and classification of COVID-19, the reporting and transfer of suspected cases, and the treatment of suspected cases. CONCLUSIONS: Primary care doctors' role perceptions in the COVID-19 pandemic are not always consistent with government guidelines in some aspects, such as transferring and diagnosing suspected cases. Therefore, it is essential to guide primary care doctors in performing their duties, especially those with lower professional titles.

8.
J Med Chem ; 65(17):11840-11853, 2022.
Article in English | Web of Science | ID: covidwho-2016520

ABSTRACT

Site-selective lysine modification of peptides and proteins in aqueous solutions or in living cells is still a big challenge today. Here, we report a novel strategy to selectively quinolylate lysine residues of peptides and proteins under native conditions without any catalysts using our newly developed water-soluble zoliniums. The zoliniums could site-selectively quinolylate K350 of bovine serum albumin and inactivate SARS-CoV-2 3CL(pro) via covalently modifying two highly conserved lysine residues (K5 and K61). In living HepG2 cells, it was demonstrated that the simple zoliniums (5b and 5B) could quinolylate protein lysine residues mainly in the nucleus, cytosol, and cytoplasm, while the zolinium-fluorophore hybrid (8) showed specific lysosome-imaging ability. The specific chemoselectivity of the zoliniums for lysine was validated by a mixture of eight different amino acids, different peptides bearing potential reactive residues, and quantum chemistry calculations. This study offers a new way to design and develop lysine-targeted covalent ligands for specific application.

9.
Journal of Xi'an Jiaotong University (Medical Sciences) ; 43(5):658-662, 2022.
Article in Chinese | EMBASE | ID: covidwho-2010481

ABSTRACT

In the emergency of the outbreak of COVID-19 in December 2019, Shaanxi Provincial Health Committee mobilized several medical teams from major hospitals in the province, and, by relying on Xi'an Chest Hospital, jointly established an anti COVID-19 consortium to control and eradicate the epidemic in a short time. Information support is an important guarantee for winning this battle. In order to realize the efficient cooperation among multiple medical teams, we have carried out some exploratory and innovative information support services on the basis of the original information system of the chest hospital. In this process, we have gone through some detours. Some compromises were made on some problems that could not be solved in the short term. Finally, in an environment full of uncertainty, a set of information support management system with basically smooth operation was built through rapid trial and error adjustment. The system mainly includes the following aspects: support of the organizational structure and operation process of the anti-epidemic consortium, support for medical collaboration related businesses of multiple medical teams, and support for statistical reports and online meetings. Information support has played a very important role in this action, and this practice has also accumulated experience for us to deal with similar situations in the future.

10.
Journal of virology ; : e0102422, 2022.
Article in English | MEDLINE | ID: covidwho-2008764

ABSTRACT

Zoonotic coronaviruses represent an ongoing threat to public health. The classical porcine epidemic diarrhea virus (PEDV) first appeared in the early 1970s. Since 2010, outbreaks of highly virulent PEDV variants have caused great economic losses to the swine industry worldwide. However, the strategies by which PEDV variants escape host immune responses are not fully understood. Complement component 3 (C3) is considered a central component of the three complement activation pathways and plays a crucial role in preventing viral infection. In this study, we found that C3 significantly inhibited PEDV replication in vitro, and both variant and classical PEDV strains induced high levels of interleukin-1β (IL-1β) in Huh7 cells. However, the PEDV variant strain reduces C3 transcript and protein levels induced by IL-1β compared with the PEDV classical strain. Examination of key molecules of the C3 transcriptional signaling pathway revealed that variant PEDV reduced C3 by inhibiting CCAAT/enhancer-binding protein β (C/EBP-β) phosphorylation. Mechanistically, PEDV nonstructural protein 1 (NSP1) inhibited C/EBP-β phosphorylation via amino acid residue 50. Finally, we constructed recombinant PEDVs to verify the critical role of amino acid 50 of NSP1 in the regulation of C3 expression. In summary, we identified a novel antiviral role of C3 in inhibiting PEDV replication and the viral immune evasion strategies of PEDV variants. Our study reveals new information on PEDV-host interactions and furthers our understanding of the pathogenic mechanism of this virus. IMPORTANCE The complement system acts as a vital link between the innate and the adaptive immunity and has the ability to recognize and neutralize various pathogens. Activation of the complement system acts as a double-edged sword, as appropriate levels of activation protect against pathogenic infections, but excessive responses can provoke a dramatic inflammatory response and cause tissue damage, leading to pathological processes, which often appear in COVID-19 patients. However, how PEDV, as the most severe coronavirus causing diarrhea in piglets, regulates the complement system has not been previously reported. In this study, for the first time, we identified a novel mechanism of a PEDV variant in the suppression of C3 expression, showing that different coronaviruses and even different subtype strains differ in regulation of C3 expression. In addition, this study provides a deeper understanding of the mechanism of the PEDV variant in immune escape and enhanced virulence.

11.
PLoS ONE [Electronic Resource] ; 17(8):e0273344, 2022.
Article in English | MEDLINE | ID: covidwho-2002328

ABSTRACT

This study explored the roles of epidemic-spread-related behaviors, vaccination status and weather factors during the COVID-19 epidemic in 50 U.S. states since March 2020. Data from March 1, 2020 to February 5, 2022 were incorporated into panel model. The states were clustered by the k-means method. In addition to discussing the whole time period, we also took multiple events nodes into account and analyzed the data in different time periods respectively by panel linear regression method. In addition, influence of cluster grouping and different incubation periods were been discussed. Non-segmented analysis showed the rate of people staying at home and the vaccination dose per capita were significantly negatively correlated with the daily incidence rate, while the number of long-distance trips was positively correlated. Weather indicators also had a negative effect to a certain extent. Most segmental results support the above view. The vaccination dose per capita was unsurprisingly proved to be the most significant factor especially for epidemic dominated by Omicron strains. 7-day was a more robust incubation period with the best model fit while weather had different effects on the epidemic spread in different time period. The implementation of prevention behaviors and the promotion of vaccination may have a successful control effect on COVID-19, including variants' epidemic such as Omicron. The spread of COVID-19 also might be associated with weather, albeit to a lesser extent.

12.
Liver Int ; 2022.
Article in English | PubMed | ID: covidwho-2001716

ABSTRACT

BACKGROUND AND AIMS: Chronic liver disease (CLD) patients and liver transplant (LT) recipients have an increased risk of morbidity and mortality from coronavirus disease 2019 (COVID-19). The immunogenicity of COVID-19 vaccines in CLD patients and LT recipients is poorly understood. The present study aimed to evaluate the immunogenicity of COVID-19 vaccines in CLD patients and LT recipients. METHODS: We searched electronic databases for eligible studies. Two reviewers independently conducted the literature search, extracted the data, and assessed the risk of bias and the quality of included studies. The rates of detectable immune response were pooled from single-arm studies. For comparative studies, we compared the rates of detectable immune response between patients and healthy controls. The meta-analysis was conducted using the Stata software with a random-effects model. RESULTS: In total, 19 observational studies involving 4191 participants met the inclusion criteria. The pooled rates of detectable humoral immune response after two doses of COVID-19 vaccination in CLD patients and LT recipients were 95% (95% confidence interval (CI)=88%-99%) and 66% (95% CI=57%-74%), respectively. After two doses of vaccination, the humoral immune response rate was similar in CLD patients and healthy controls (risk ratio (RR)=0.96;95% CI=0.90-1.02;P=0.14). In contrast, LT recipients had a lower humoral immune response rate after two doses of vaccination than healthy controls (RR=0.68;95% CI=0.59-0.77;P<0.01). CONCLUSIONS: Our meta-analysis demonstrated that COVID-19 vaccination induced strong humoral immune responses in CLD patients but poor humoral immune responses in LT recipients.

13.
Nat Nanotechnol ; 2022.
Article in English | PubMed | ID: covidwho-2000903

ABSTRACT

The global emergency caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic can only be solved with effective and widespread preventive and therapeutic strategies, and both are still insufficient. Here, we describe an ultrathin two-dimensional CuInP(2)S(6) (CIPS) nanosheet as a new agent against SARS-CoV-2 infection. CIPS exhibits an extremely high and selective binding capacity (dissociation constant (K(D)) < 1 pM) for the receptor binding domain of the spike protein of wild-type SARS-CoV-2 and its variants of concern, including Delta and Omicron, inhibiting virus entry and infection in angiotensin converting enzyme 2 (ACE2)-bearing cells, human airway epithelial organoids and human ACE2-transgenic mice. On association with CIPS, the virus is quickly phagocytosed and eliminated by macrophages, suggesting that CIPS could be successfully used to capture and facilitate virus elimination by the host. Thus, we propose CIPS as a promising nanodrug for future safe and effective anti-SARS-CoV-2 therapy, and as a decontamination agent and surface-coating material to reduce SARS-CoV-2 infectivity.

14.
30th IEEE/ACM International Symposium on Quality of Service, IWQoS 2022 ; 2022.
Article in English | Scopus | ID: covidwho-1992651

ABSTRACT

With the growing interest in web services during the current COVID-19 outbreak, the demand for high-quality low-latency interactive applications has never been more apparent. Yet, packet losses are inevitable over the Internet, since it is based on UDP. In this paper, we propose Ivory, a new real-world system framework designed to support network adaptive error control in real-time communications, such as VoIP, using a recently proposed low-latency streaming code. We design and implement our prototype over UDP that can correct or retransmit lost packets conditional on network conditions and application requirements.To maintain the highest quality, Ivory attempts to correct as many lost packets as possible on-the-fly, yet incurring the smallest footprint in terms of coding overhead over the network. To achieve such an objective, Ivory uses a deep reinforcement learning agent to estimate the best coding parameters in real-time based on observed network states and experience learned. It learns offline the best coding parameters to use based on previously observed loss patterns and takes into account the round-trip time observed to decide on the optimum decoding delay for a low-latency application. Our extensive array of experiments shows that Ivory achieves a better trade-off between recovering packets and using lower redundancy than the state-of-the-art network adaptive streaming codes algorithms. © 2022 IEEE.

16.
Zhonghua Liu Xing Bing Xue Za Zhi ; 43(5): 655-662, 2022 May 10.
Article in Chinese | MEDLINE | ID: covidwho-1969571

ABSTRACT

2019-nCoV Omicron (B.1.1.529) variant, which has brought new challenges to the prevention and control of COVID-19 pandemic, has the characteristics of stronger transmissibility and more rapid transmission and more significant immune evasion. It took only two months to become a predominant strain worldwide after its identification in South Africa in November 2021. Local epidemics caused by Omicron variant have been reported in several provinces in China. However, the epidemiological characteristics of highly mutated Omicron variant remain unclear. This article summarizes the progress in the research of functional mutations, transmissibility, virulence, immune evasion and cross-reactive immune responses of Omicron variant, to provide references for the effective prevention and control of COVID-19 pandemic caused by Omicron variant.


Subject(s)
COVID-19 , SARS-CoV-2 , Humans , Mutation , Pandemics
17.
Journal of Uncertain Systems ; 2022.
Article in English | Scopus | ID: covidwho-1962387

ABSTRACT

Under the background of economic globalization and professional division of labor, each link of enterprise supply chain management is facing more and more risks. Over the past two years, due to the turbulent situation at home and abroad and the repeated outbreaks of COVID-19 around the world, the normal procurement work has been greatly affected which means the purchase cost increases and even out of stock occur. The rapid development of big data, artificial intelligence and other technologies has brought new tools and means for enterprise risk management. This paper focuses on the procurement work in enterprise operation management, and analyzes how to obtain procurement related news or current review texts, and analyzes their emotional tendency to evaluate and quantify the public opinion risk of procurement by natural language processing (NLP), denoted as NLP. By combing the crawled text data, we add some purchase related words to a corpus which can analyze good and bad emotions and train a model. We use the model to score the manually labeled text data to determine the optimal threshold of positive news and negative news. Under this threshold, the accuracy of text emotion analysis is 85.4%. Finally, through a case analysis, we show the specific implementation of procurement public opinion risk score evaluation. © 2022 World Scientific Publishing Company.

18.
2021 International Conference on Computer Application and Information Security, ICCAIS 2021 ; 12260, 2022.
Article in English | Scopus | ID: covidwho-1932601

ABSTRACT

COVID-19 plays role in every part of the world;especially, it does harm to lives of people. Thus, COVID-19 sounds the alarm that is very important to build an effective mechanism to help prevent pandemic disease. In this work, dynamic network based on status value is built, which aims to help simulate the added danger level by the addition of infected people or close contacts. First, each node of this network is labelled with different kinds of status which has special value to show its danger degree. Then, the weight of the network represents the relationship of nodes;with the value of each node, average length and average spread of danger level is calculated based on the accumulation of dynamic weight. Thus, epidemic speed and scope of the infectious disease can be simulated. Moreover, the experiments compared to other networks have verified the effectiveness of our model. © The Authors.

19.
23rd IEEE International Conference on High Performance Computing and Communications, 7th IEEE International Conference on Data Science and Systems, 19th IEEE International Conference on Smart City and 7th IEEE International Conference on Dependability in Sensor, Cloud and Big Data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021 ; : 1245-1252, 2022.
Article in English | Scopus | ID: covidwho-1909206

ABSTRACT

In this work, we study national and state-level COVID-19 pandemic data in the United States with the help of human mobility trend data and auxiliary medical information. We analyze and compare various state-of-the-art time-series prediction techniques. We assess a spatio-temporal graph neural network model which forecasts the pandemic course by utilizing a hybrid deep learning architecture and human mobility data. Nodes in the graph represent the state-level deaths due to COVID-19 at any particular time point, edges represent the human mobility trend and temporal edges correspond to node attributes across time. We also study statistical modeling and machine learning techniques for mortality prediction in the United States. We evaluate these techniques on both state and national level COVID-19 data in the United States and claim that the SARIMAX and GCN-LSTM model generated forecast values using exogenous hospital information variables can enrich the underlying model to improve the prediction accuracy at both levels. Our best machine learning models perform 50% and 60% better than the baseline on an average on the national level and state-level data, respectively, while the statistical models perform 63% and 42% better. © 2021 IEEE.

20.
American Journal of Respiratory and Critical Care Medicine ; 205:1, 2022.
Article in English | English Web of Science | ID: covidwho-1880679
SELECTION OF CITATIONS
SEARCH DETAIL