An observer comparison study to evaluate a machine learning model to quantify the infected pneumonia on lung CT images

Quantification of infected lung volume using computed tomography (CT) images can play a critical role in predicting the severity of pulmonary infectious disease. Manual segmentation of infected areas from several CT image slices, however, is not efficient and viable in clinical practice. To assist clinicians in overcoming this challenge, we developed a new method to automatically segment and quantify the percentage of the infected lung volume. First, we used a public dataset of 20 COVID-19 patients, which consists of manually annotated lung and infection masks, to train a new joint deep learning (DL) model for lung and infection segmentation. As for lung segmentation, a Mask-RCNN model was applied to the lung volume with a novel postprocessing technique. Following that, an ensemble model with a customized residual attention UNet model and feature pyramid network (FPN) models was employed for infection segmentation. Next, we assembled another set of 80 CT scans of Covid-19 patients. Two chest radiologists manually evaluated each CT scan and reported the infected lung volume percentage using a customized graphical user interface (GUI). The developed DL-model was also employed to process these CT images. Then, we compared the agreement between the radiologist (manual) and model-based (automated) percentages of diseased regions. Additionally, the GUI was used to let radiologists rate acceptance of the DL-model generated segmentation results. Analyzing the results demonstrate that the agreement between manual and automated segmentation is >95% in 28 testing cases. Furthermore, >53% of testing cases received the top assessment rating scores from two radiologists (between four-five- score). Thus, this study illustrates the feasibility of developing a DL-model based automated tool to effectively provide quantitative evaluation of infected lung regions to assist in improving the efficiency of radiologists in infection diagnosis. © COPYRIGHT SPIE. Downloading of the is permitted for personal use only.

[Latent period and incubation period with associated factors of COVID-19 caused by Omicron variant].

Objective: To estimate the latent period and incubation period of Omicron variant infections and analyze associated factors. Methods: From January 1 to June 30, 2022, 467 infections and 335 symptomatic infections in five local Omicron variant outbreaks in China were selected as the study subjects. The latent period and incubation period were estimated by using log-normal distribution and gamma distribution models, and the associated factors were analyzed by using the accelerated failure time model (AFT). Results: The median (Q1, Q3) age of 467 Omicron infections including 253 males (54.18%) was 26 (20, 39) years old. There were 132 asymptomatic infections (28.27%) and 335 (71.73%) symptomatic infections. The mean latent period of 467 Omicron infections was 2.65 (95%CI: 2.53-2.78) days, and 98% of infections were positive for nucleic acid test within 6.37 (95%CI: 5.86-6.82) days after infection. The mean incubation period of 335 symptomatic infections was 3.40 (95%CI: 3.25-3.57) days, and 97% of them developed clinical symptoms within 6.80 (95%CI: 6.34-7.22) days after infection. The results of the AFT model analysis showed that compared with the group aged 18-49 years old, the latent period [exp(ß)=1.36 (95%CI: 1.16-1.60), P<0.001] and incubation period [exp(ß)=1.24 (95%CI: 1.07-1.45), P=0.006] of infections aged 0-17 years old were prolonged. The latent period [exp(ß)=1.38 (95%CI: 1.17-1.63), P<0.001] and the incubation period [exp(ß)=1.26 (95%CI: 1.06-1.48), P=0.007] of infections aged 50 years old and above were also prolonged. Conclusion: The latent period and incubation period of most Omicron infections are within 7 days, and age may be a influencing factor of the latent period and incubation period.

Safety and Pharmacokinetics of a Novel Double Prodrug Asc10 in Healthy Subjects

Background: ASC10 is an oral double prodrug of the active antiviral ribonucleoside analog, ASC10-A (also known as beta-d-N4-hydroxycytidine), which is a potent inhibitor of SARS-CoV-2. ASC10 is rapidly metabolized into ASC10-A in vivo after oral dosing. Here, we report the results of the first-in-human, phase 1 study to determine the safety, tolerability, and pharmacokinetics (PK) of ASC10 in healthy subjects, and to assess the food effect on the pharmacokinetics. Method(s): This study included 2 parts. Part 1 (multiple-ascending-dose) consisted of 6 cohorts (8 or 12 subjects per cohort). Eligible subjects were randomized in a 3:1 ratio to receive either twice-daily (BID) doses of 50 to 800 mg ASC10 or placebo for 5.5 days, and were then followed for 7 days for safety. In Part 2 (food effect), 12 subjects were randomized in a 1:1 ratio to either 800 mg ASC10 in the fed state followed by 800 mg in the fasted state, or vice versa, with a 7-day washout period between doses. PK blood samples were collected and measured for ASC10-A along with ASC10 and molnupiravir. Safety assessments included monitoring of adverse events (AEs), measurement of vital signs, clinical laboratory tests, and physical examinations. Result(s): ASC10-A was the major circulating metabolite ( >99.94%) in subjects after oral dosing of ASC10. ASC10-A appeared rapidly in plasma, with a median Tmax of 1.00 to 2.00 h, and declined with a geometric t1/2 of approximately 1.10 to 3.04 h. After multiple dosing for 5.5 days, both Cmax and AUC of ASC10-A increased in a dose-proportional manner from doses of 50 to 800 mg BID without accumulation. of ASC10-A in the fed state occurred slightly later, with a median of 3.99 h postdose versus 2.00 h (fasted state). However, Cmax and AUC were very similar or the same between fed and fasted states. Thus, administration of ASC10 with food is unlikely to have an effect on exposure. The incidence of AEs was similar between subjects receiving ASC10 or placebo (both 66.7%) and 95.0% of AEs were mild. There were no serious adverse events as well as no clinically significant findings in clinical laboratory, vital signs, or electrocardiography. Conclusion(s): Results of this study showed that ASC10 was well tolerated, and the increase in plasma exposure of ASC10-A was dose proportional across the range of doses tested with no accumulation and no food effect. 800 mg ASC10 BID is selected for further studies in patients infected with SARS-CoV-2.

Understanding Hospitality and Tourism Students' Emotional Intelligence Performance in the E-learning Environment: A Delphi Approach

The COVID 19 pandemic has forced educators and students to embrace e-learning. It has become urgent that educators expedite their efforts in establishing criteria to assess the overall effectiveness of e-learning, in which student emotional intelligence (EI) cultivation and development play an increasingly centric role. However, a survey of the current literature shows that EI in e-learning appears to have received little attention. This study was thus designed to help fill this research void. Specifically, it set out to understand typical hospitality and tourism students' EI behaviors in the e-learning environment. To achieve this goal, this study applied a two-round Delphi approach. The findings show that in the e-learning environment, students commonly exhibit high self-awareness, low self-management, low social management, and low relationship-building competence. Prior EI studies mainly focus on employee performance and behavior but this study extends the effect of EI in education and offers significant implications for hospitality and tourism educators and researchers (word count: 155). © 2022 ICHRIE.

Research on The Index System of Laboratory Assets Digitization

Under the influence of the fifth industrial revolution and the outbreak of COVID-19, the digital transformation of enterprises has entered a new stage of rapid development. Digital transformation has become the trend of enterprise operation in the digital economy era. In this context, enterprise laboratory asset operation has also become an important aspect of enterprise digital operation. It is urgent to build a set of enterprise laboratory asset digital evaluation system to assist the implementation of enterprise digital strategy. Based on the characteristics of laboratory assets and the closed-loop theory of asset operation management, this paper analyzes and studies the laboratory asset management, establishes a targeted evaluation index system of digital asset management, focuses on the composition of the digital operation system of laboratory assets, and constructs a management index evaluation system of assets, efficiency, cost and other dimensions, so as to create a real-time, comprehensive and comprehensive evaluation system The closed-loop and full cycle digital management ecological environment realizes the effective integration of laboratory resource fragmentation information and the complete embodiment of digitization, provides service support for continuously improving asset management performance, and provides support for further improving enterprise economic efficiency and operation level. © 2022 SPIE.

Effect of E-beam irradiation on the qualitative attributes of shrimp (Penaeus vannamei)

As a non-thermal food processing technology, Electron beam (E-beam) irradiation has been used to enhance microbial safety by deactivating unwanted spoilage and pathogenic microorganisms in food industry. This study evaluated the effects of E-beam irradiation at doses killing SARS-COV-2 on qualities and sensory attributes. The results showed that irradiation caused little effect on the proximate composition, amino acid content, texture, and sensory attributes (P > 0.05). However, E-beam increased TBARS (Thiobarbituric acid reactive substances) and lowered vitamin E content in dose-dependently. Irradiation up to 10 kGy significantly decreased unsaturated fatty acid (UFA) content and inhibited the increase in TVB-N (The total volatile basic nitrogen) while reducing cohesiveness and chewiness (P < 0.05). E-beam irradiation with 7–10 kGy caused greater ΔE values (ΔE > 5) via the significant increase of b*, accompanied by big visual difference in shrimp (P < 0.05). A dose of 4 kGy E-beam irradiation was recommended without altering its physicochemical properties and sensory attributes. © 2023 Elsevier Ltd

Research on The Index System of Laboratory Assets Digitization

Under the influence of the fifth industrial revolution and the outbreak of COVID-19, the digital transformation of enterprises has entered a new stage of rapid development. Digital transformation has become the trend of enterprise operation in the digital economy era. In this context, enterprise laboratory asset operation has also become an important aspect of enterprise digital operation. It is urgent to build a set of enterprise laboratory asset digital evaluation system to assist the implementation of enterprise digital strategy. Based on the characteristics of laboratory assets and the closed-loop theory of asset operation management, this paper analyzes and studies the laboratory asset management, establishes a targeted evaluation index system of digital asset management, focuses on the composition of the digital operation system of laboratory assets, and constructs a management index evaluation system of assets, efficiency, cost and other dimensions, so as to create a real-time, comprehensive and comprehensive evaluation system The closed-loop and full cycle digital management ecological environment realizes the effective integration of laboratory resource fragmentation information and the complete embodiment of digitization, provides service support for continuously improving asset management performance, and provides support for further improving enterprise economic efficiency and operation level. © 2022 SPIE.

The Impact of Government Behavior on the Development of Cross-Border E-Commerce B2B Export Trading Enterprises Based on Evolutionary Game in the Context of "Dual-Cycle" Policy

In 2020, the COVID-19 pandemic had a major impact on China's foreign trade. Therefore, the Chinese government has proposed a "dual cycle" policy to promote economic development. In 2021, China's cross-border e-commerce B2B exports accounted for 60 percent. Therefore, this paper studies the impact of government actions on the development of cross-border e-commerce B2B export enterprises under the background of "dual cycle" policy. First, the policies related to the cross-border e-commerce industry in the "dual circulation" policy are screened, and the LDA topic model is used to classify them, i.e., sorting by topic intensity as "fiscal policy", "tax policy", "customs clearance policy", "payment policy" and "talent policy". After that, based on the analysis results of the LDA topic model, a theoretical basis for the impact of different policies on cross-border e-commerce B2B export companies is established;then an evolutionary game model between the government and cross-border e-commerce B2B export enterprises is constructed. This article also carried out experiments to verify our analysis. The simulation results show that: (1) The government's appropriate increase in subsidies, tax incentives, infrastructure investment, talent introduction and cultivation, optimized payment system, and supervision can promote enterprises to participate in cross-border e-commerce B2B export trading;(2) excessive government supervision reduces enterprises' enthusiasm to participate in cross-border e-commerce B2B export trading;(3) the government's subsidies, tax incentives, and supervision strength have the greatest impact on whether enterprises participate in cross-border e-commerce B2B export trading, followed by the government's investment in cross-border e-commerce infrastructure, the introduction and cultivation of cross-border e-commerce talents, and the improvement of the payment system. Finally, this paper puts forward relevant policy recommendations to promote the development of cross-border e-commerce B2B export enterprises.

Covid-19 vaccination can induce multiple sclerosis via cross-reactive CD4+ T cells recognizing SARS-CoV-2 spike protein and myelin peptides

Introduction: Infection with the SARS-CoV-2 coronavirus can lead to a wide range of acute and also chronic disease manifestations. The rapidly developed vaccinations are highly effective in preventing severe disease courses and have been proven safe. Both natural infection and, to a much lower extent, the mRNAbased vaccinations can be accompanied by transient autoimmune phenomena or onset of autoimmune diseases. Objective(s): We report here two cases of multiple sclerosis (MS) with clinical and new radiological signs beginning in close temporal relation to spike (S) protein mRNA-based vaccinations. Aim(s): To establish that the onset of MS in these two cases is very likely caused by CD4+ T cell clones that cross-recognize SARSCoV- 2 S protein-derived peptides and peptides derived from myelin proteins, which have previously been implicated in MS. Method(s): Spike specific CD4+ T cells from peripheral blood and CD4+ T cells from CSF sample were isolated and expanded for autoantigen screening test. A list of well-known MS-related autoantigens including immunodominant peptides and isoforms from MBP, MOG, PLP, RASGRP2, TSTA3 peptides were included to assess T cell reactivity. CD4+ CFSElow fraction were sorted after stimulate with positive autoantigen pools or SARSCov- 2 Spike protein, followed by expansion and testing with autoantigen peptides and Spike protein. Supernatant from cell culture were further analyzed for IFN-gamma secretion. Result(s): Self-reactive T cells were detected from Spike specific T cell population in both patients. CD4+ T from CSF also showed reactivity to MBP, MOG, PLP peptide pools. Finally, we found proinflammatory T cell clones that recognize both Spike protein and immunodominant MBP peptides and MOG peptides, which have previously been implicated in MS. Conclusion(s): Detailed studies of both peripheral blood- and CSFderived CD4+ T cells show that the onset of MS in these two cases is very likely caused by CD4+ T cell clones that cross-recognize SARS-CoV-2 S protein-derived peptides and peptides derived from myelin proteins, which have previously been implicated in MS.

Prediction and Analysis of Infectious Diseases Based on M-SIR Model

The epidemic of infectious diseases has become a major problem threatening the world public health, and the dynamic models of virus spreading are widely used for epidemic tracking and prediction. The existing dynamic models do not consider the synergistic effects of population migration factors and changes in transmission rates on diseases. Therefore, based on the SIR (Susceptible-Infectious-Recovered) model, the time-dependent M-SIR (Migration-Susceptible-Infectious-Recovered) model was proposed by introducing the population migration (Migration) factor. Meanwhile, introducing the machine learning LightGBM (Light Gradient Boosting Machine) method to track the infection rate and recovery rate, and explored the impact of cross-regional population movement and prevention and control measures on the development of the epidemic. Take the new crown epidemic as an example, firstly, the data of population migration and epidemic spread were statistically analyzed to monitor the relationship between population mobility and epidemic development. Then, the m-sir model is used to predict the infected cases and removed cases in Beijing and Shanghai. Through comparative analysis with the SIR model, the prediction accuracy of the model has been greatly improved. At the same time, the development trend of the epidemic situation in related cities before and after control is explored, which can provide some theoretical support for future epidemic prediction and control decisions. © 2022 IEEE.

Understanding Hospitality and Tourism Students’ Emotional Intelligence Performance in the E-learning Environment: A Delphi Approach

The COVID 19 pandemic has forced educators and students to embrace e-learning. It has become urgent that educators expedite their efforts in establishing criteria to assess the overall effectiveness of e-learning, in which student emotional intelligence (EI) cultivation and development play an increasingly centric role. However, a survey of the current literature shows that EI in e-learning appears to have received little attention. This study was thus designed to help fill this research void. Specifically, it set out to understand typical hospitality and tourism students’ EI behaviors in the e-learning environment. To achieve this goal, this study applied a two-round Delphi approach. The findings show that in the e-learning environment, students commonly exhibit high self-awareness, low self-management, low social management, and low relationship-building competence. Prior EI studies mainly focus on employee performance and behavior but this study extends the effect of EI in education and offers significant implications for hospitality and tourism educators and researchers (word count: 155). © 2022 ICHRIE.

Epidemiology and evolution of novel deltacoronaviruses in birds in central China

BIRDS : The variety and widespread of coronavirus in natural reservoir animals is likely to cause epidemics via interspecific transmission, which has attracted much attention due to frequent coronavirus epidemics in recent decades. Birds are natural reservoir of various viruses, but the existence of coronaviruses in wild birds in central China has been barely studied. Some bird coronaviruses belong to the genus of Deltacoronavirus. To explore the diversity of bird deltacoronaviruses in central China, we tested faecal samples from 415 wild birds in Hunan Province, China. By RT-PCR detection, we identified eight samples positive for deltacoronaviruses which were all from common magpies, and in four of them, we successfully amplified complete deltacoronavirus genomes distinct from currently known deltacoronavirus, indicating four novel deltacoronavirus stains (HNU1-1, HNU1-2, HNU2 and HNU3). Comparative analysis on the four genomic sequences showed that these novel magpie deltacoronaviruses shared three different S genes among which the S genes of HNU1-1 and HNU1-2 showed 93.8% amino acid (aa) identity to that of thrush coronavirus HKU12, HNU2 S showed 71.9% aa identity to that of White-eye coronavirus HKU16, and HNU3 S showed 72.4% aa identity to that of sparrow coronavirus HKU17. Recombination analysis showed that frequent recombination events of the S genes occurred among these deltacoronavirus strains. Two novel putative cleavage sites separating the non-structural proteins in the HNU coronaviruses were found. Bayesian phylogeographic analysis showed that the south coast of China might be a potential origin of bird deltacoronaviruses existing in inland China. In summary, these results suggest that common magpie in China carries diverse deltacoronaviruses with novel genomic features, indicating an important source of environmental coronaviruses closed to human communities, which may provide key information for prevention and control of future coronavirus epidemics

DETERMINING THE BEST ACOUSTIC FEATURES FOR SMOKER IDENTIFICATION

Speech-based automatic smoker identification (also known as smoker/non-smoker classification) aims to identify speakers' smoking status from their speech. In the COVID-19 pandemic, speech-based automatic smoker identification approaches have received more attention in smoking cessation research due to low cost and contactless sample collection. This study focuses on determining the best acoustic features for smoker identification. In this paper, we investigate the performance of four acoustic feature sets/representations extracted using three feature extraction/learning approaches: (i) hand-crafted feature sets including the extended Geneva Minimalistic Acoustic Parameter Set and the Computational Paralinguistics Challenge Set, (ii) the Bag-of-Audio-Words representations, (iii) the neural representations extracted from raw waveform signals by SincNet. Experimental results show that: (i) SincNet feature representations are the most effective for smoker identification and outperform the MFCC baseline features by 16% in absolute accuracy;(ii) the performance of hand-crafted feature sets and the Bag-of-Audio-Words representations rely on the scale of the dimensions of feature vectors. © 2022 IEEE

Extraction-free RT-PCR to Detect SARS-CoV-2 Variants of Concern

Background. SARS-CoV-2 variants of concern (VOC) have challenged real-time reverse transcriptase polymerase chain reaction (RT-PCR) methods for the diagnosis of COVID-19. Methods. The CDC 2019-Novel Coronavirus real-time RT-PCR panel was modified to create a single-plex extraction-free proxy RT-PCR assay, VOCFast™. This assay uses the nucleocapsid N1 as well as novel primer/probe pairs to target VOC mutations in the Orf1a and spike (S) genes. For analytical validation of VOCFast, synthetic controls for the Wuhan, alpha/B.1.1.7, beta/B.1.351, and gamma/P.1 strains were tested at various concentrations. Clinical validation was performed using patient anterior nares swab and saliva specimens collected in the Denver, CO area between Nov 2020 and Feb 2021 or in March 2021. Orthogonal next-generation sequencing (NGS) was also performed. Results. Similar N1 quantification cycle (Cq) values corresponding to viral load were observed for all strains, suggesting that VOC mutations do not affect performance of the N1 primer/probe. Orf1a-mut and S1-mut primer/probes generated a stable high Cq value for the Wuhan strain. Conversely, Orf1a-mut Cq values were inversely correlated with viral load for all VOC. The S1-mut Cq was inversely correlated with viral load of the alpha strain, but did not reliably amplify beta/gamma VOC. The limit of detection was 8 copies/uL. The first set of COVID-19 patient specimens revealed no amplification using Orf1amut whereas 53% of specimens collected in Mar 2021 demonstrated amplification by Orf-1a. Orthogonal testing by the SARS-CoV-2 NGS Assay and COVID-DX software demonstrated that 12/12 alpha strains, 2/2 beta/gamma strains, and 33/33 Wuhan strains were correctly identified by VOCFast. Conclusion. The combination of the N1, Orf1a-mut, and S1-mut primers/probes in VOCFast can distinguish the Wuhan, alpha, and beta/gamma strains and it consistent with NGS results. Testing of clinical samples revealed that VOC emerged in Denver, CO in March 2021. Future work to discriminate beta, gamma, and emerging VOC is ongoing. In summary, VOCFast is an extraction-free RT-PCR assay for nasal swab and saliva specimens that can identify VOC with a turnaround time suitable for clinical testing.

A Study on the Relationship between Online Learning and Emotional Quotient of College Students during the Period of Epidemic Prevention and Control

Background: The outbreak of the COVID-19 forced school education from offline to online, and then the effectiveness of online teaching has become a close concern of educators. According to the interview, college students were easy to be emotional instability during the time of online learning, which made their life rhythms disordered, academic performances declined as well as bad mood appeared frequently. Hence, this paper mainly discusses the relationship between online teaching and emotional quotient of college students during the period of epidemic prevention and control. Methods: Based on the interviews with college students, the questionnaire was designed by EIS, IRI-C and Teachers' personality test. We took the college students as Subjects of investigation, and collected 372 effective electronic questionnaires nationally. 12 invalid questionnaires were deleted,372 valid questionnaires remained, and the effective recovery rate was 96.88%. Factor analysis and Hierarchical regression were performed with SPSS 19.0 and Excel to test the correlation and moderating effect. Results: (1) There is a strong correlation between online learning and emotional quotient of college students during the period of epidemic prevention and control. The five main factors that affect the emotional quotient of college students are self-control ability, discernment, interpersonal relationship, self-motivation, teachers' character. (2) The influence of teachers' character on students' interpersonal relationship has an indirect effect on college students' emotional quotient. (3) Self-control ability, discernment, interpersonal relationship, self-motivation, teachers' character and the emotional quotient of college students are in good condition. College students have strong empathy, but low self-control ability. Conclusions: The emotional quotient of college students is affected by internal and external factors. Therefore, to improve the emotional quotient level of college students, we need to pay attention to not only internal factors, but also external factors. © 2021 IEEE.

Remote video-based suturing education with smartphones (REVISE): a randomized controlled trial

Background: Suturing is a skill that surgical trainees must master. Owing to the COVID-19 pandemic, medical schools needed to transition from in-person to virtual avenues of teaching, including procedural skills. Our primary objective was to determine whether virtual video-based feedback is no worse than in-person feedback in improving novice medical students' suturing skills. Methods: Fifty-four medical students were randomly assigned either to an experimental arm in which they received remote-recorded feedback (RRF) or remote-live feedback (RLF) or to a control arm in which they received in-person feedback (control). There were 18 participants in each group. Participants first learned to suture via an online module then recorded themselves performing a standardized suturing task at home. Customized feedback was then provided by a surgical resident who received standardized training for this project. Results: RRF participants received a feedback video, RLF participants received live feedback over Zoom and control participants received feedback in person. Participants then recorded another video of the same suturing task. Prefeedback and postfeedback suturing performances were scored by blinded assessors using the University of Bergen Suturing Skills Assessment Tool (UBAT). Our primary outcome measure is the score difference between pre- and post-feedback videos. The RLF and RRF groups were compared for statistical significant differences using a 2-tailed paired t test. Results: Twenty-seven participants (median age 22) were included in the interim analyses. Postfeedback UBAT scores were not significantly different between groups (70.34 [interquartile range (IQR) -25.41 to 127.34] v. 11.34 [IQR -99.66 to 78.34], p > 0.05), with a higher score indicating better performance. Although there was a trend toward RLF demonstrating greater improvement (32 [IQR -10 to 59.25] v. 25 [IQR -63 to 77], p > 0.05), this was not statistically significant. Conclusion: Thus far, there has been no significant difference between groups in prefeedback scores, postfeedback scores or score difference. Future steps include analyzing videos for all participants and comparison with the control arm.

Evaluation of the effectiveness of EFL online teaching during the COVID-19 pandemic

Online teaching has been massively conducted during the novel coronavirus period all over the world. How to evaluate online teaching has been increasingly researched recently. This study looked at how English as a foreign language (EFL) teaching was delivered online by university teachers during the COVID-19 pandemic. We investigated university teachers and students’ perception of effective EFL online teaching and learning based on several evaluation modes in using technology in education. Data were collected using questionnaires and interviews from teachers and students in a variety of provinces in Mainland China. The results showed that various methods were used to deliver online EFL courses and these approaches are found to correlate with each other. Teachers and students provided positive comments on online teaching and were satisfied with their online teaching and learning. Participants also noted effective ways in online EFL teaching. The findings indicated that when teachers have more training, more skills, and more confidence, they could deliver more effective online teaching and learning. © The Author(s) 2021.

A novel Nongfu Spring's best selling strategy based on big data

In order to overcome the threat of traditional drinking water in marketing, such as fierce market competition, threats of new entrants, and achieving leading enterprise, this paper proposes a novel Nongfu Spring's best selling strategy based on big data. Except the pricing solutions such as the discount on bundles or boxes, sales signs, 8-endings, and premium price, the tactics are different from traditional ones in that they focus on reinforcement of a 'natural and healthy' brand image, improvement of diversity and creativity in product design, and higher investment in direct and social marketing. The research results show that those tactics could assist increase health awareness after the COVID-19 pandemic, build a better brand image, satisfy a variety of customer needs, increase the scope of the promotion campaign, and more importantly, maintain dominant market share of Nongfu Spring. © 2020 IEEE.

MiniSeg: An Extremely Minimum Network for Efficient COVID-19 Segmentation

The rapid spread of the new pandemic, i.e., COVID-19, has severely threatened global health. Deep-learning-based computer-aided screening, e.g., COVID-19 infected CT area segmentation, has attracted much attention. However, the publicly available COVID-19 training data are limited, easily causing overfitting for traditional deep learning methods that are usually data-hungry with millions of parameters. On the other hand, fast training/testing and low computational cost are also necessary for quick deployment and development of COVID-19 screening systems, but traditional deep learning methods are usually computationally intensive. To address the above problems, we propose MiniSeg, a lightweight deep learning model for efficient COVID-19 segmentation. Compared with traditional segmentation methods, MiniSeg has several significant strengths: i) it only has 83K parameters and is thus not easy to overfit;ii) it has high computational efficiency and is thus convenient for practical deployment;iii) it can be fast retrained by other users using their private COVID-19 data for further improving performance. In addition, we build a comprehensive COVID-19 segmentation benchmark for comparing MiniSeg to traditional methods.