Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 20 de 43
Filter
1.
Open Forum Infectious Diseases ; 9(Supplement 2):S451-S452, 2022.
Article in English | EMBASE | ID: covidwho-2189722

ABSTRACT

Background. COVID-19 pandemic, especially during resurgences of cases in hard-hit areas, led to significant shortage of hospital beds. Such shortages may be alleviated through timely and effective forecasting of hospital discharges. The objective of this study is to predict next 7-day discharges of hospitalized COVID-19 patients using daily-based electronic health records (EHR) data. Methods. Using EHR data of hospitalized COVID-19 patients from 03/2020-08/ 2021, we employed ensemble learning to predict next 7-day discharges of individual patients. We used both baseline and daily inpatient features for model training, validation, and test. Baseline features include demographic and clinical characteristics, and comorbidities. The daily inpatient features were vital signs, laboratory tests, medications administered, acute physiological scores, use of ventilator, and use of intensive care unit. 1832 hospitalized patients were identified (12,397 hospital days). Samples were randomly split at patient level (7:2:1) into training set (N=1,283 patients with 8,704 hospital days), validation set (N=366 patients with 2,524 days), and test/ holdout set (patient N=183, and 1,169 days). Prediction models were trained on the training set and the validation set. We conducted the model training separately on the samples of admission day and the samples of days after admission day. The predictions were based on the ensemble learning from decision tree, XGBoost, logistic regression, and multilayer perceptron, long short-term memory (LSTM), bi-directional LSTM, and convolutional neural network. The combination of ensemble learning on the test/holdout set was used for final next 7-day predictions based on 'hard' voting (by majority). Where there was a tie, we used 'soft' voting (sum of probabilities) to break the tie. (Figure Presented) Results. The overall average hospital length of stay was 8.7 (SD=10.5) days. The ensemble learning accuracies for admission-day samples and after-admission-day samples were 0.781 and 0.793, and the F1-scores for were 0.761 and 0.789, respectively. Conclusion. EHR data of hospitalized COVID-19 patients can be used to predict next 7-day hospital discharges. Additional inpatient features and more advanced machine learning techniques are needed for prediction accuracy improvement.

2.
Acs Nano ; 11:11, 2023.
Article in English | MEDLINE | ID: covidwho-2185519

ABSTRACT

Plasmonic metasurfaces (PMs) functionalized with the monoclonal antibody (mAb) are promising biophotonic sensors for biomolecular interaction analysis and convenient immunoassay of various biomarkers, such as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants. Previous PM biosensing suffers from the slow affinity detection rate and lack of sufficient immunoassay studies on various SARS-CoV-2 variants. Here, we develop a high-efficiency affinity testing method based on label-free PM sensors with mAbs and demonstrate their binding characteristics to 12 spike receptor binding domain (RBD) variants of SARS-CoV-2. In addition to the research of plasmonic near-field influence on surface biomolecule sensing, we provide a comprehensive report about the Langmuir binding equilibrium of molecular kinetics between 12 SARS-CoV-2 RBD variants and mAb-functionalized PMs, which plays a crucial role in label-free immunosensing. A high-affinity mAb can be combined with the highly sensitive propagating plasmonic mode to boost the detection of SARS-CoV-2 variants. Owing to a better understanding of molecular dynamics on PMs, we develop an ultrasensitive biosensor of the SARS-CoV-2 Omicron variant. The experiments show great distinguishment of P < 0.0001 from respiratory diseases induced by other viruses, and the limit of detection is 2 orders smaller than the commercial colloidal gold immunoassay. Our study shows the label-free biosensing by low-cost wafer-scale PMs, which will provide essential information on biomolecular interaction and facilitate high-precision point-of-care testing for emerging SARS-CoV-2 variants in the future.

3.
Parasitology ; 2022.
Article in English | EMBASE | ID: covidwho-2185291

ABSTRACT

The apicomplexan parasite Cyclospora cayetanensis causes seasonal foodborne outbreaks of the gastrointestinal illness cyclosporiasis. Prior to the COVID-19 pandemic, annually reported cases were increasing in the United States (US), leading the US Centers for Disease Control and Prevention (CDC) to develop a genotyping tool to complement cyclosporiasis outbreak investigations. Thousands of US isolates and one from China (strain CHN-HEN01) were genotyped by Illumina amplicon sequencing, revealing two lineages (A and B). The allelic composition of isolates was examined at each locus. Two nuclear loci (CDS3 and 360i2) distinguished lineages A and B. CDS3 had two major alleles;one almost exclusive to lineage A and the other to lineage B. Six 360i2 alleles were observed-two exclusive to lineage A (alleles A1 and A2), three to lineage B (B1 and B2), and one (B4) was exclusive to CHN-HEN01 which shared allele B3 with lineage B. Examination of heterozygous genotypes revealed that mixtures of A-And B-Type 360i2 alleles occurred rarely, suggesting a lack of gene flow between lineages. Phylogenetic analysis of loci from whole genome shotgun sequences, mitochondrial, and apicoplast genomes, revealed that CHN-HEN01 represents a distinct lineage (C). Retrospective examination of epidemiologic data revealed associations between lineage and the geographical distribution of US infections plus strong temporal associations. Given the multiple lines of evidence for speciation within human-infecting Cyclospora, we provide an updated taxonomic description of C. cayetanensis, and describe two novel species as etiological agents of human cyclosporiasis;Cyclospora ashfordi sp. nov. and Cyclospora henanensis sp. nov. (Apicomplexa: Eimeriidae). Copyright © 2022 Cambridge University Press. All rights reserved.

4.
Frontiers in Artificial Intelligence ; 5:1034732, 2022.
Article in English | MEDLINE | ID: covidwho-2199578

ABSTRACT

Since 2019, the COVID-19 pandemic has had an extremely high impact on all facets of the society and will potentially have an everlasting impact for years to come. In response to this, over the past years, there have been a significant number of research efforts on exploring approaches to combat COVID-19. In this paper, we present a survey of the current research efforts on using mobile Internet of Thing (IoT) devices, Artificial Intelligence (AI), and telemedicine for COVID-19 detection and prediction. We first present the background and then present current research in this field. Specifically, we present the research on COVID-19 monitoring and detection, contact tracing, machine learning based approaches, telemedicine, and security. We finally discuss the challenges and the future work that lay ahead in this field before concluding this paper.

5.
Advances in Production Engineering & Management ; 17(4):425-438, 2022.
Article in English | ProQuest Central | ID: covidwho-2204004

ABSTRACT

With the gradual normalization of the COVID-19, unmanned delivery has gradually become an important contactless distribution method around China. In this paper, we study the routing problem of unmanned vehicles considering path flexibility and the number of traffic lights in the road network to reduce the complexity of road conditions faced by unmanned vehicles as much as possible. We use Monte Carlo Tree Search algorithm to improve the Genetic Algorithm to solve this problem, first use Monte Carlo Tree Search Algorithm to compute the time-saving path between two nodes among multiple feasible paths and then transfer the paths results to Genetic Algorithm to obtain the final sequence of the unmanned vehicles fleet. And the hybrid algorithm was tested on the actual road network data around four hospitals in Beijing. The results showed that compared with normal vehicle routing problem, considering path flexibility can save the delivery time, the more complex the road network composition, the better results could be obtained by the algorithm.

6.
Chinese Traditional and Herbal Drugs ; 54(1):192-209, 2023.
Article in Chinese | EMBASE | ID: covidwho-2203149

ABSTRACT

Objective To analyze the medication rules of related epidemic disease prescription in Treatise on Febrile Diseases based on data mining, and the mechanism of "Chaihu (Bupleuri Radix)-Huangqin (Scutellariae Radix)" as the core drugs in the treatment of coronavirus disease 2019 (COVID-19) by network pharmacology, in order to explore the contemporary value of classical prescriptions in the treatment of epidemic diseases. Methods The prescriptions for treating epidemic diseases in Treatise on Febrile Diseases were screened, and the medication rules such as drug frequency, flavor and meridian tropism as well as correlation, apriori algorithm were analyzed by using software such as R language. The mechanism of the core drugs in the medication pattern in the treatment of COVID-19 was explored by the network pharmacology. A "disease-drug-ingredient-target" network was constructed on the selected components and targets with Cytoscape. The key targets were introduced into String database for network analysis of protein-protein interaction (PPI), and gene ontology (GO) functional analysis and Kyoto encyclopedia of genes and genomes (KEGG) pathway analysis were conducted in R language. Results A total of 61 prescriptions for treating epidemic diseases in Treatise on Febrile Diseases were included, including 52 traditional Chinese medicines (TCMs). In the top 20 high-frequency drugs, warm drugs, spicy drugs and qitonifying drugs were mainly used, mostly in the spleen and lung meridian. Chaihu (Bupleuri Radix) and Huangqin (Scutellariae Radix) herb pair had the strongest correlation. A total of five clusters were excavated: supplemented formula of Xiaochaihu Decoction (), Sini Decoction (), supplemented formule of Maxing Shigan Decoction (), Fuling Baizhu Decoction () and Dachengqi Decoction (). A total of 45 active ingredients, 189 action targets of Bupleuri Radix-Scutellariae Radix herb pair, and 543 targets of COVID-19 were obtained from TCMSP and Genecards, and 64 intersection targets were generated. The results of the network analysis showed that the main components of core drugs pair against COVID-19 may be quercetin, wogonin, kaempferol baicalein, acacetin etc., and the core targets may be VEGFA, TNF, IL-6, TP53, AKT1, CASP3, CXCL8, PTGS2, etc. A total of 1871 related entries and 164 pathways were obtained by GO and KEGG enrichment analysis, respectively. Conclusion In Treatise on Febrile Diseases, the treatment of epidemic diseases mainly chose pungent, warm, spleen-invigorating and qi-tonifying herbs, such as Xiaochaihu Decoction, Sini Decoction and Dachengqi Decoction, etc. It was found that Bupleuri Radix-Scutellariae Radix core herb pair prevent and treat COVID-19 through multi-target targets such as PTGS2, IL-6 and TNF. The ancient prescriptions for treating epidemic disease in Treatise on Febrile Diseases may have significant reference value for the prevention and treatment of new epidemic diseases today. Copyright © 2023 Editorial Office of Chinese Traditional and Herbal Drugs. All rights reserved.

7.
Clin Chem Lab Med ; 2022.
Article in English | PubMed | ID: covidwho-2154344

ABSTRACT

OBJECTIVES: Various comorbidities associated with COVID-19 add up in severity of the disease and obviously prolonged the time for viral clearance. This study investigated a novel ultrasensitive MAGLUMI(®) SARS-CoV-2 Ag chemiluminescent immunoassay assay (MAG-CLIA) for diagnosis and monitoring the infectivity of COVID-19 patients with comorbid conditions during the pandemic of 2022 Shanghai. METHODS: Analytical performances of the MAG-CLIA were evaluated, including precision, limit of quantitation, linearity and specificity. Nasopharyngeal specimens from 232 hospitalized patients who were SARS-CoV-2 RT-qPCR positive and from 477 healthy donors were included. The longitudinal studies were performed by monitoring antigen concentrations alongside with RT-qPCR results in 14 COVID-19 comorbid participants for up to 22 days. The critical antigen concentration in determining virus infectivity was evaluated at the reference cycle threshold (Ct) of 35. RESULTS: COVID-19 patients were well-identified using an optimal threshold of 0.64 ng/L antigen concentration, with sensitivity and specificity of 95.7% (95% CI: 92.2-97.9%) and 98.3% (95% CI: 96.7-99.3%), respectively, while the Wondfo LFT exhibited those of 34.9% (95% CI: 28.8-41.4%) and 100% (95% CI: 99.23-100%), respectively. The sensitivity of MAG-CLIA remained 91.46% (95% CI: 83.14-95.8%) for the samples with Ct values between 35 and 40. Close dynamic consistence was observed between MAG-CLIA and viral load time series in the longitudinal studies. The critical value of 8.82 ng/L antigen showed adequate sensitivity and specificity in evaluating the infectivity of hospitalized convalescent patients with comorbidities. CONCLUSIONS: The MAG-CLIA SARS-CoV-2 Ag detection is an effective and alternative approach for rapid diagnosis and enables us to evaluate the infectivity of hospitalized convalescent patients with comorbidities.

8.
BMC Nephrol ; 23(1):389, 2022.
Article in English | PubMed | ID: covidwho-2153529

ABSTRACT

BACKGROUND: Observational studies have shown home hemodialysis (HHD) to be associated with better survival than facility hemodialysis (HD) and peritoneal dialysis (PD). Patients on HHD have reported higher quality of life and independence. HHD is considered to be an economical way to manage end-stage kidney disease (ESKD). The coronavirus disease 2019 pandemic has had a significant impact on patients with ESKD. Patients on HHD may have an advantage over in-center HD patients because of a lower risk of exposure to infection. PARTICIPANTS AND METHODS: We enrolled HD patients from our dialysis center. We first established the HHD training center. The training center was approved by the Chinese government. Doctors, nurses and engineers train and assess patients separately. There are three forms of patient monitoring: home visits, internet remote monitoring, and outpatient services. Demographic and medical data included age, sex, blood pressure, and dialysis-related data. Laboratory tests were conducted in our central testing laboratory, including hemoglobin (Hgb), serum creatinine (Cr), urea nitrogen (BUN), uric acid (UA), albumin (Alb), calcium (Ca), phosphorus (P), parathyroid hormone (PTH), and brain natriuretic peptide (BNP) levels. RESULTS: Six patients who underwent regular dialysis in the HD center of our hospital were selected for HHD training. We enrolled 6 patients, including 4 males and 2 females. The mean age of the patients was 47.5 (34.7-55.7) years, and the mean dialysis age was 33.5 (11.2-41.5) months. After an average of 16.0 (11.2-25.5) months of training, Alb, P and BNP levels were improved compared with the baseline values. After training, three patients returned home to begin independent HD. During the follow-up, there were no serious adverse events leading to hospitalization or death, but there were several adverse events. They were solved quickly by extra home visits of the technicians or online by remote monitoring. During the follow-up time, the laboratory indicators of all the patients, including Hgb, Alb, Ca, P, PTH, BNP, and β2-MG levels, remained stable before and after HHD treatment. CONCLUSION: HHD is feasible and safe for ESKD in China, but larger-scale and longer-term studies are needed for further confirmation.

9.
Jisuanji Gongcheng/Computer Engineering ; 48(8), 2022.
Article in Chinese | Scopus | ID: covidwho-2145860

ABSTRACT

In recent years, the COVID-19, which involves a highly infectious virus, has spread worldwide.Wearing masks in public areas can reduce the transmission and hence the spread of the virus.Additionally, using computer vision technology to detect mask wearing behavior in public areas is crucial.To prevent and control epidemics, the correct form of wearing face masks must be identified.In an actual environment, the detection of mask wearing is complex and diverse.The scale of a face wearing a mask is different;furthermore, the difference between the correct and wrong forms of wearing a mask is subtle and hence difficult to detect.Therefore, a mask wearing detection algorithm based on an improved Single Shot Multibox Detector(SSD) algorithm is proposed herein.Based on the SSD network, the algorithm introduces a feature fusion network and an attention coordination mechanism, reconstructs the feature extraction network, and enhances the ability of learning and processing detailed information.In addition, the classification prediction score and IoU score of the algorithm are combined, whereas the Quality Focal Loss(QFL) function is used to adjust the weight of positive and negative samples.An experiment is performed on acustom-developed mask wearing test dataset.Experimental results show that the average accuracy of the algorithm is 96.28%, which is 5.62% higher than that of the original algorithm.The improved algorithm offers good accuracy and practicability for mask wearing detection, as well assatisfies the requirements for epidemic prevention and control. © 2022, Editorial Office of Computer Engineering. All rights reserved.

10.
Zhonghua Liu Xing Bing Xue Za Zhi ; 43(11): 1699-1704, 2022 Nov 10.
Article in Chinese | MEDLINE | ID: covidwho-2143855

ABSTRACT

Objective: To clarify the epidemiological characteristics and spatiotemporal clustering dynamics of COVID-19 in Shanghai in 2022. Methods: The COVID-19 data presented on the official websites of Municipal Health Commissions of Shanghai during March 1, 2022 and May 31, 2022 were collected for a spatial autocorrelation analysis by GeoDa software. A logistic growth model was used to fit the epidemic situation and make a comparison with the actual infection situation. Results: Pudong district had the highest number of symptomatic and asymptomatic infectants, accounting for 29.30% and 35.58% of the total infectants. Differences in cumulative attack rates and infection rates among 16 districts (P<0.001) were significant. The rates were significantly higher in Huangpu district than in other districts. The attack rate of COVID-19 from March 1, 2022 to May 31, 2022 had a global spatial positive correlation (P<0.05). Spatial distribution of COVID-19 attack rate was different at different periods. The global autocorrelation coefficient from March 16 to March 29, April 6 to April 12 and May 18 to May 24 had no statistical significance (P>0.05). Our local autocorrelation analysis showed that 22 high-high clustering areas were detected in eight periods.The high-risk hot-spot areas have experienced a "less-more-less" change process. The growth model fitting results were consistent with the actual infection situation. Conclusion: There was a clear spatiotemporal correlation in the distribution of COVID-19 in Shanghai. The comprehensive prevention and control measures of COVID-19 epidemic in Shanghai have effectively prohibited the growth of the epidemic, not only curbing the spatially spread of high-risk epidemic areas, but also reducing the risk of transmission to other cities.


Subject(s)
COVID-19 , Epidemics , Humans , COVID-19/epidemiology , China/epidemiology , Disease Outbreaks , Spatial Analysis
11.
IEEE Transactions on Computational Social Systems ; : 1-13, 2022.
Article in English | Scopus | ID: covidwho-2097656

ABSTRACT

Since the outbreak of COVID-19, an alternative way to keep students on the track, meanwhile, prevent them from being at the risk of infection is in highly demand. Many education providers had made a move in trial of delivering knowledge and learning materials remotely. Along with this trend, learning management systems, open educational resources (OERs) and OER platforms, mini applications in social media and video-conference software were combined in a rush to create a multi-channel delivery mode to make learning resources openly available round-the-clock. Learning activities in this fast migration to online were regularly found to be carried out in gradual and fragmented time spans. Due to the little-known learner information along with the continuously released new OERs, the cold start problem still hinders the innovative mode of delivery and adaptive micro learning. To overcome the data sparsity, an online computation is proposed to benefit OER providers and instructors. A lightweight learner-micro-OER profile and two algorithmic solutions are provided to tackle the new user and new item cold start problem, respectively. Learning paths are generated and optimized in terms of heuristic rules to form the initial recommendation list. By adopting the same set of rules, newly released micro OERs are inserted into established learning paths to increase their discoverability. IEEE

12.
23rd Annual Conference of the International Speech Communication Association, INTERSPEECH 2022 ; 2022-September:2168-2172, 2022.
Article in English | Scopus | ID: covidwho-2091312

ABSTRACT

A fast, efficient and accurate detection method of COVID-19 remains a critical challenge. Many cough-based COVID-19 detection researches have shown competitive results through artificial intelligence. However, the lack of analysis on vocalization characteristics of cough sounds limits the further improvement of detection performance. In this paper, we propose two novel acoustic features of cough sounds and a convolutional neural network structure for COVID-19 detection. First, a time-frequency differential feature is proposed to characterize dynamic information of cough sounds in time and frequency domain. Then, an energy ratio feature is proposed to calculate the energy difference caused by the phonation characteristics in different cough phases. Finally, a convolutional neural network with two parallel branches which is pre-trained on a large amount of unlabeled cough data is proposed for classification. Experiment results show that our proposed method achieves state-of-the-art performance on Coswara dataset for COVID-19 detection. The results on an external clinical dataset Virufy also show the better generalization ability of our proposed method. Copyright © 2022 ISCA.

13.
International Conference on Green Building, Civil Engineering and Smart City, GBCESC 2022 ; 211 LNCE:465-473, 2023.
Article in English | Scopus | ID: covidwho-2059767

ABSTRACT

The COVID-19 pandemic has seen the importance of confined space ventilation to reduce the risks of cross infection. To evaluate and compare the relative impacts of different mitigation strategies is important in order to reduce the risk of infection in a given situation. Using CFD methods, this study aimed to modulate the spread of exhaled contaminants in a floor-heated and ventilated space. Three different inlet velocities and four floor temperatures were used to assess the effect of the radiant floor combined with the displacement ventilation (RFDV) on room airflow and pollutant spread. Results show that RFDV reduced exposure to infection from 87% to 50% compared to the reference case. The inlet velocity is required to increase when the floor temperature is higher to decrease the contaminant exposure risk to in the room. This research provides a timely and necessary study of the ventilation and heating systems. These findings are expected to be useful for designing future of RFDV. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

15.
Mathematical Biosciences and Engineering ; 19(12):11854-11867, 2022.
Article in English | Web of Science | ID: covidwho-2006289

ABSTRACT

Infectious diseases generally spread along with the asymmetry of social network propagation because the asymmetry of urban development and the prevention strategies often affect the direction of the movement. But the spreading mechanism of the epidemic remains to explore in the directed network. In this paper, the main effect of the directed network and delay on the dynamic behaviors of the epidemic is investigated. The algebraic expressions of Turing instability are given to show the role of the directed network in the spread of the epidemic, which overcomes the drawback that undirected networks cannot lead to the outbreaks of infectious diseases. Then, Hopf bifurcation is analyzed to illustrate the dynamic mechanism of the periodic outbreak, which is consistent with the transmission of COVID-19. Also, the discrepancy ratio between the imported and the exported is proposed to explain the importance of quarantine policies and the spread mechanism. Finally, the theoretical results are verified by numerical simulation.

16.
FRONTIERS IN EDUCATION ; 7, 2022.
Article in English | Web of Science | ID: covidwho-1938611

ABSTRACT

Since the coronavirus disease 2019 (COVID-19) pandemic, human parasitology education has been exceedingly disrupted. To deliver human parasitology knowledge, medical universities in China have employed multiple measures, some of which have had positive outcomes that have not yet been summarized. The objective of this review is to share the Chinese experience as the human parasitology teaching methods were transformed. In general, we adopted a fully online teaching model under urgent pandemic control measures based on a well-structured teaching model that integrated the course preview, live lecture, review, and assessment. Combinations were attempted of COVID-19 and parasitology teaching contents. Some active learning models, such as case-based e-learning and flipped classrooms, were proposed for offline and online blended teaching during the normalization stage of the pandemic. Meanwhile, we discuss both the strengths and flaws of online and blended teaching. Some useful assessment tools are presented for reference purposes. In conclusion, this transition to online and online-offline blended human parasitology teaching in China has boosted innovative teaching activities and may continue to catalyze the transformation of medical education.

17.
EPL ; 137(4), 2022.
Article in English | Scopus | ID: covidwho-1846890

ABSTRACT

It is well known that the outbreak of infectious diseases is affected by the diffusion of the infected. However, the diffusion network is seldom considered in the network-organized SIR model. In this work, we investigate the effect of the maximum eigenvalue on Turing instability and show the role of network parameters (the network connection rate, the network's infection, etc.) on the outbreak of infectious diseases. Meanwhile, stability of network-organized SIR is given by using the maximum eigenvalue of the network matrix which is proportional to the network connection rate and the networks infection rate. The bridge between the two rates and Turing instability was also revealed which can explain the spread mechanism of infectious diseases. In the end, some measures to mitigate the spread of infectious diseases are proposed and the feasible strategies for prevention and control can be provided in our paper, the data from COVID-19 validated the above results. © Copyright © 2022 EPLA.

18.
Educational Technology and Society ; 25(1):75-77, 2022.
Article in English | Scopus | ID: covidwho-1728468

ABSTRACT

COVID-19 pandemic had changed the world-wide education landscape as the whole society is adapting to the “new normal.” We orgainised a special issue collecting research papers that shed insights on how teaching and learning designs will be affected, and how novel educational technologies will help in a fast post-pandemic recovery. 26 papers were received but only 11 papers were finally selected to publish, after two rounds of rigorous reviews. This editorial note discusses the background, quality management and thematic topic groups of the papers. © 2022,Educational Technology and Society. All rights reserved

19.
3rd International Conference on Modeling, Simulation, Optimization and Algorithm, ICMSOA 2021 ; 2173, 2022.
Article in English | Scopus | ID: covidwho-1701138

ABSTRACT

In this study, we established a SEQR model for the spread of COVID-19. The impact of city lockdown measures and other factors on the spread of the epidemic was discussed in this paper through dynamic analysis and numerical simulation of the model. The research results showed that the input ratio of the exposed population should be controlled to at least 0.001 if the government wanted to effectively control the epidemic. Closing the city was the most effective way. The exposed population of COVID-19 in Xiaogan City was hoping to be cleared on February 29 under strict control. The closed city could be released around March 28. But at the same time, we should pay attention to prevent the virus from entering it again. The peak number of existing patients would reduce by about 1,400 if city lockdown measures were taken in Xiaogan City four days in advance. © Published under licence by IOP Publishing Ltd.

20.
2nd International Symposium on Artificial Intelligence for Medicine Sciences, ISAIMS 2021 ; : 209-215, 2021.
Article in English | Scopus | ID: covidwho-1613107

ABSTRACT

Cough-based COVID-19 detection has shown competitive results through artificial intelligence. In this paper, we proposed a cough-based COVID-19 detection method that made full use of frequency information at different stage. In the feature extraction stage, we proposed band weighted Mel-Frequency Cepstral Coefficients to emphasize features at different frequency bands;in the classification stage, we proposed a multi-band Long-Short Term Memory Convolutional Neural Network with attention mechanism to capture detailed features in the frequency domain. We also combined SpecAugment and Mixup to improve the generalization ability of our proposed model. We evaluated the performance of our proposed model on the dataset of DiCOVA challenge 2021 and our collected dataset. Experimental results showed that the AUC of our model outperformed the first place of DiCOVA challenge 2021 by 5.11% absolutely on average. © 2021 ACM.

SELECTION OF CITATIONS
SEARCH DETAIL