Flexible work arrangements and gender differences in research during the COVID-19 period in Zimbabwean higher learning institutions

This study sought to investigate the impact of COVID-19-induced flexible work arrangements (FWAs) on gender differences in research outputs during COVID-19. A mixed research methodology was used, focusing on higher learning institutions in Zimbabwe. Purposive sampling was applied to select 250 researchers from the 21 registered universities in Zimbabwe. The study's findings revealed that institutions of higher learning in Zimbabwe did not provide the necessary affordances to enable both male and female academics to work from home effectively. The study also established that FWAs were preferred and appreciated by both male and female academics. However, whilst both male and female academics performed their teaching responsibilities without incident, unlike males, females struggled to find time for research, thus affecting professional growth and development negatively for female academics. Cultural traditions were found to subordinate females to domestic and caregiving responsibilities unrelated to their professions. The findings raise questions on the feasibility of the much-recommended FWAs for future work on female academics' research careers. Thus, without the necessary systems and processes to support female researchers, FWAs can only widen the gender gap in research outputs. This study contributes to the Zimbabwean higher learning institutions' perspective on how FWAs' policies and practices could be re-configured to assist female researchers in enhancing their research outputs as well as their career growth.

AUTOANTIBODIES RELATED TO RHEUMATIC DISEASES AFTER COVID-19 INFECTION

BackgroundSARS-CoV-2(Severe acute respiratory syndrome coronavirus 2) has been circulating worldwide for three years. It mainly causes upper respiratory tract infection, which can manifest as pulmonary infection and even respiratory distress syndrome in severe cases. Different autoantibodies can be detected in patients infected with COVID-19.ObjectivesTo explore autoantibodies related to rheumatic diseases after COVID-19 infection.MethodsNinety-eight inpatients were tested for antinuclear antibodies (ANA), antibodies to extractable nuclear antigens(ENA), anti-neutrophil cytoplasmic antibodies（ANCA), anticardiolipin antibodies,a-β2GPI (IgG/IgM). They were from a tertiary hospital in Guangzhou during the COVID-19 epidemic. Data were described statistically.ResultsNinety-eight hospitalized patients were tested for relevant antibodies. The average age was 50.64±19.54;67 (68.4%) were male, 64 (65.3%) were COVID-19 positive, 90 (90.9%) had rheumatic diseases, and 56 of them were COVID-19 positive patients with rheumatic diseases.There were 76 patients tested for antinuclear antibodies;29 （38.16%）were negative, 18 （23.68%）had a 1/80 titre, and 29（28.16%） had a titre greater than 1:80. The 31 covid patients were positive for ANA. In the high-titer group, 19 patients with rheumatic diseases were positive for COVID-19, and 12 patients had an exacerbation of the rheumatic diseases (6 of whom had previously had pulmonary fibrosis). Of 31 covid patients, only two were non-rheumatic patients, and both were elderly, aged 85 and 100, respectively.Fifty-six patients had ENA results, and 29 for positive antibodies, 8 for ds-DNA antibodies, 2 for anti-Sm antibodies, 6 for anti-nucleosome antibodies, 12 for anti-U1RNP antibodies, 2 for anti-Scl-70 antibodies, 12 for anti-SS-A antibodies, 3 for anti-mitochondrial M2 antibodies, 2 for anti-centromere antibodies, 1 for anti-Po antibodies, and one for anti-Jo-1 antibody. All 56 patients had rheumatic diseases, and no new patients were found.There were 62 patients with ANCA data. P-ANCA was positive in 12 cases(19.35%), and MPO-ANCA was positive in 2 cases. An 85-year-old non-rheumatic COVID-19 patient was P-ANCA positive. She had a history of hypertension, colon cancer, CKD3, coronary heart disease, and atrial flutter.In the anticardiolipin antibodies group, there were 62 patients;only 6 were positive, and 2 were rheumatic patients infected with COVID-19. Antiphospholipid antibodies were detected in 33 patients, and a-β2GPI was tested in one patient, an 82-year-old COVID-19 patient with gout, diabetes, and cerebral infarction in the past. We did not find a statistical difference in the above results.ConclusionWe have not found a correlation between SARS-CoV-2 and serum autoantibodies of rheumatic immune diseases. It needs large samples and an extended follow-up to research.AcknowledgementsThis work was supported by Scientific and Technological Planning Project of Guangzhou City [202102020150], Guangdong Provincial Basic and Applied Basic Research Fund Project [2021A1515111172], National Natural Science Foundation of China Youth Fund [82201998] and Third Affiliated Hospital of Sun Yat-Sen University Cultivating Special Fund Project for National Natural Science Foundation of China [2022GZRPYQN01].Disclosure of Interestsone declared.

A Mask Detection Algorithm Based on Improved Yolov5s

Aiming at the severe form of new coronavirus epidemic prevention and control, a target detection algorithm is proposed to detect whether masks are worn in public places. The Ghostnet and SElayer modules with fewer design parameters replace the BottleneckCSP part in the original Yolov5s network, which reduces the computational complexity of the model and improves the detection accuracy. The bounding box regression loss function DIOU is optimized, the DGIOU loss function is used for bounding box regression, and the center coordinate distance between the two bounding boxes is considered to achieve a better convergence effect. In the feature pyramid, the depthwise separable convolution DW is used to replace the ordinary convolution, which further reduces the amount of parameters and reduces the loss of feature information caused by multiple convolutions. The experimental results show that compared with the yolov5s algorithm, the proposed method improves the mAP by 4.6% and the detection rate by 10.7 frame/s in the mask wearing detection. Compared with other mainstream algorithms, the improved yolov5s algorithm has better generalization ability and practicability. © 2022 ACM.

Spatial Interactions and the Spread of COVID-19: A Network Perspective

We use unique data on the travel history of confirmed patients at a daily frequency across 31 provinces in China to study how spatial interactions influence the geographic spread of pandemic COVID-19. We develop and simultaneously estimate a structural model of dynamic disease transmission network formation and spatial interaction. This allows us to understand what externalities the disease risk associated with a single place may create for the entire country. We find a positive and significant spatial interaction effect that strongly influences the duration and severity of pandemic COVID-19. And there exists heterogeneity in this interaction effect: the spatial spillover effect from the source province is significantly higher than from other provinces. Further counterfactual policy analysis shows that targeting the key province can improve the effectiveness of policy interventions for containing the geographic spread of pandemic COVID-19, and the effect of such targeted policy decreases with an increase in the time of delay.

METHODOLOGY for ASSESSING the FINANCIAL SECURITY of ENTERPRISES in the POST-PANDEMIC PERIOD of DIGITAL ECONOMY

The article considers the approaches to assessing the financial security of enterprises presented in the literature, determines the rsistance of the textile industry of Uzbekistan to the negative impact of the coronavirus pandemic on the basis of statistical data, and reveals a significant differentiation of textile industry enterprises in terms of financial stability. Based on data on small enterprises in the textile industry of Uzbekistan, a method for assessing the financial security of an enterprise in the post-pandemic period is proposed and tested, taking into account the complex influence of non-financial parameters of economic security and assessing the deviations of the economic situation at a given enterprise from the patterns emerging in the relevant segment of the economy. In this research an econometric model was developed to determine the factors affecting the chemical industry and express their interrelationship, based on the conducted econometric analysis, the directions of development in our country were determined. According to the authors, it is necessary to continue these directions in order to ensure the economic security of industry enterprises in the country. © 2022 ACM.

Coping Orientation of Academic Community in the Time of COVID-19 Pandemic: A Pilot Multi-Country Survey Study

In this paper, we have mapped the coping methods used to address the coronavirus pandemic by members of the academic community. We conducted an anonymous survey of a convenient sample of 674 faculty/staff members and students from September to December 2020. A modified version of the RCOPE scale was used for data collection. The results indicate that both religious and existential coping methods were used by respondents. The study also indicates that even though 71% of informants believed in God or another religious figure, 61% reported that they had tried to gain control of the situation directly without the help of God or another religious figure. The ranking of the coping strategies used indicates that the first five methods used by informants were all non-religious coping methods (i.e., secular existential coping methods): regarding life as a part of a greater whole, regarding nature as an important resource, listening to the sound of surrounding nature, being alone and contemplating, and walking/engaging in any activities outdoors giving a spiritual feeling. Our results contribute to the new area of research on academic community's coping with pandemic-related stress and challenges.

RHEUMATOID ARTHRITIS FLARE FOLLOWING COVID-19 VACCINATION IN MALAYSIAN POPULATION AND ITS ASSOCIATED RISK FACTORS

BackgroundFlare of Rheumatoid Arthritis (RA) following COVID-19 vaccination has been reported with a low occurrence observed in those patients with disease remission. However, no local data is available in our multi-ethnic Malaysian population.ObjectivesTo evaluate the prevalence of RA flare in Malaysian patients following COVID-19 vaccination and its associated risk factors.MethodsThis was a cross-sectional study assessing RA flare based on patient-reported disease flare through self-administered questionnaires and physician-reported flare. Patient self-reported disease flare was defined as ‘a sudden worsening of rheumatology condition or arthritis within 1 month post-vaccination' while physician-reported flare was defined as ‘an increment of disease activity score 28-joint documented within 3 months post-vaccination‘ from either a scheduled or unscheduled clinic visit. A total of 186 RA patients attended the rheumatology clinic in Hospital Putrajaya from May to July 2022 who completed the primary COVID-19 vaccination under the Malaysian National Vaccination Programme were recruited. Demographic data, disease parameters including serology for rheumatoid factor (RF) and anti-citrullinated peptide antibodies (ACPA), cessation of disease modifying anti-rheumatic drugs (DMARDs) around vaccination, type of vaccines and adverse events were examined using descriptive and univariate analyses.ResultsMajority (93%) of RA patients enrolled were female with a mean age of 58 years old (standard deviation, SD 12.2) and mean disease duration was 12 years (SD 7.7). More than half were seropositive (66% RF, 63% ACPA) with 47.4% had double seropositivity (RF and ACPA positive). All patients received DMARDs with the majority (71%) were on methotrexate (MTX), 21.5% were on leflunomide, 17.7% on other DMARDs, with a small proportion (14%) of patients were receiving prednisolone. Only 4.8% of patients were on biologics or targeted synthetic disease modifying anti-rheumatic drugs. Half of the patients were in remission prior to vaccination. 62% of patients received Pfizer-BioNTech vaccine as the primary vaccine, followed by Sinovac-CoronaVac (24.6%) and Oxford-AstraZeneca (13.4%) vaccines. A booster dose had been administered to 80% of patients, of which 88.7% was Pfizer-BioNTech vaccine. MTX therapy were discontinued in 39.4% of patients (n=52) post-vaccination for a week duration. The prevalence of RA flare was only 12.9% (n=24) in which 14 were self-reported and 10 were physician-reported flares (4 severe flare, 6 mild-moderate flare). Flare rates were higher during the first and second dose of vaccination with 29.2% respectively, and only 12.5% were reported after booster vaccination. Common vaccine adverse effects were fever (16.8%), myalgia (8.6%) and arthralgia (6.4%). There were no significant differences in the occurrence of flare post-vaccination between age, gender, disease activity prior to vaccination, types of vaccine, usage of MTX and prednisolone, and discontinuation of MTX post-vaccination. Although seropositivity did not exhibit statistically significant flare rate post vaccination, sub-analysis revealed four times higher rate of flare in those who has double positivity compared to seronegative RA patients (12% vs 4%).ConclusionPrevelance of RA flare post-COVID-19 vaccination in Malaysian RA population is low. No significant associated risk factors were identified although double seropositivity appeared to have higher number of flares.References[1]Bixio, R., Bertelle, D., Masia, M., Pistillo, F., Carletto, A. and Rossini, M. (2021), Incidence of Disease Flare After BNT162b2 Coronavirus Disease 2019 Vaccination in Patients With Rheumatoid Arthritis in Remission. ACR Open Rheumatology, 3: 832-833.[2]Li X, Tong X, Yeung WWY, Kuan P, Yum SHH, Chui CSL, Lai FTT, Wan EYF, Wong CKH, Chan EWY, Lau CS, Wong ICK. Two-dose COVID-19 vaccination and possible arthritis flare among patients with rheumatoid arthritis in Hong Kong. Ann Rheum Dis. 2022 Apr;81(4):564-568.Acknowledgements:NIL.Disclosure of InterestsNone Declared.

A Comprehensive Review of Sound-Based Modalities for Automatic COVID-19 Detection using Deep Learning-Based Techniques

The World Health Organization has labeled the novel coronavirus illness (COVID-19) a pandemic since March 2020. It's a new viral infection with a respiratory tropism that could lead to atypical pneumonia. Thus, according to experts, early detection of the positive cases with people infected by the COVID-19 virus is highly needed. In this manner, patients will be segregated from other individuals, and the infection will not spread. As a result, developing early detection and diagnosis procedures to enable a speedy treatment process and stop the transmission of the virus has become a focus of research. Alternative early-screening approaches have become necessary due to the time-consuming nature of the current testing methodology such as Reverse transcription polymerase chain reaction (RT-PCR) test. The methods for detecting COVID-19 using deep learning (DL) algorithms using sound modality, which have become an active research area in recent years, have been thoroughly reviewed in this work. Although the majority of the newly proposed methods are based on medical images (i.e. X-ray and CT scans), we show in this comprehensive survey that the sound modality can be a good alternative to these methods, providing faster and easiest way to create a database with a high performance. We also present the most popular sound databases proposed for COVID-19 detection. © 2022 IEEE.

Detection of Multiple Structural Breaks in Large Covariance Matrices

This article studies multiple structural breaks in large contemporaneous covariance matrices of high-dimensional time series satisfying an approximate factor model. The breaks in the second-order moment structure of the common components are due to sudden changes in either factor loadings or covariance of latent factors, requiring appropriate transformation of the factor models to facilitate estimation of the (transformed) common factors and factor loadings via the classical principal component analysis. With the estimated factors and idiosyncratic errors, an easy-to-implement CUSUM-based detection technique is introduced to consistently estimate the location and number of breaks and correctly identify whether they originate in the common or idiosyncratic error components. The algorithms of Wild Binary Segmentation for Covariance (WBS-Cov) and Wild Sparsified Binary Segmentation for Covariance (WSBS-Cov) are used to estimate breaks in the common and idiosyncratic error components, respectively. Under some technical conditions, the asymptotic properties of the proposed methodology are derived with near-optimal rates (up to a logarithmic factor) achieved for the estimated breaks. Monte Carlo simulation studies are conducted to examine the finite-sample performance of the developed method and its comparison with other existing approaches. We finally apply our method to study the contemporaneous covariance structure of daily returns of S&P 500 constituents and identify a few breaks including those occurring during the 2007–2008 financial crisis and the recent coronavirus (COVID-19) outbreak. An package "” is provided to implement the proposed algorithms.

TÜKETÍMDE TOPLUMSAL CÍNSÍYET: KÜRESEL BÍR SALGIN (COVID-19) SONRASI YAŞANAN DEĞÍŞÍMLER

Genellikle insanlar doğumdan itibaren toplumsal cinsiyet rollerine maruz kalarak ilerleyen yaşantılarında bu roller doğrultusunda tüketim tercihlerini ve kararlarını belirlemeye çalışmaktadır. Toplumdan topluma değişen skonusu bu roller, kişilerin vereceği tüketim kararlarında etkili bir pozisyonda yer almasını sağlayabilir. Ancak daha önce yapılmış çalışmalar, küresel bir şokun tüketimler üzerinde bireylerin benzer tüketim tercihlerinde bulunulduğunu göstermiştir. Bu bağlamda çalışmanın amacı Covid-19 salgını ile birlikte toplumun cinsiyetlere yüklediği rollere ait tüketim kalıplarının genç yetişkinler üzerindeki değişiminin ne şekilde yaşandığını ortaya koymayı amaçlamaktadır. Araştırma yöntemi olarak nitel araştırma tekniklerinden derinlemesine mülakat tekniği kullanılmıştır. Çalışma kapsamında Muğla Sıtkı Koçman Üniversitesi Íktisadi ve Ídari Bilimler Fakültesi bünyesinde yer alan bölümlerdeki (işletme, iktisat, kamu yönetimi, siyaset bilimleri ve uluslararası ilişkiler) 40 öğrenci ile bu görüşmeler gerçekleştirilmiştir. Sonuç olarak küresel salgınla birlikte toplumsal cinsiyet algısına aykırı satın alımlar gerçekleştiren katılımcıların son derece az oluşu hatta genç yetişkinlerin büyük çoğunluğunun toplumsal cinsiyete bağlı tüketim yapmadıkları ve buna karşı oldukları ortaya çıkmıştır.Alternate :People are often exposed to gender roles from birth to determine consumption preferences and decisions in line with these roles in their future lives. These roles, which have changed from society to society, can enable people to take an effective position in the decisions of consumption. However, previous studies have shown that individuals have similar consumption preferences on the consumption of a global shock.In this context, the aim of the study is to demonstrate the way that the consumption patterns of the roles that society has put into the gender, along with the Covid-19 epidemic, have changed on young adults. In-depth interview technique from qualitative research techniques was used as a method of research. These meetings were held with 40 students in the departments of the Faculty of Economics and Administrative Sciences (business, economics, public administration, political sciences and international relations) of Muğla Sitki Kocman University within the scope of the study. As a result, participants who have made procurement against the perception of gender with the global outbreak have been extremely low, and even the majority of young adults have shown that they do not consume and oppose gender-dependent consumption.

Supplements purchase intention: young consumer's perspective

PurposeThis study aims to investigate the factors that influence young consumers' purchase intention towards dietary supplements (DS) in Malaysia.Design/methodology/approachThe supplement industry in Malaysia has been growing rapidly recently due to a paradigm shift in healthcare management, from curative to prevention. Thus, it has sparked interest to conduct a study on the factors that influence young consumers' purchase intentions towards DS. With a response rate of 74.5%, a survey questionnaire was used to elicit data from 149 Gen-Y respondents who consume supplements on a regular basis.FindingsThe results revealed that the influencing factors towards purchase decisions among these Gen-Y respondents aged between 17 and 25 are based on product knowledge and product quality. The implications of these findings and directions for future research are outlined at the end of this paper.Originality/valueThis study offers empirical insights from the perspective of an emerging digital economy on the factors that influence young consumers' purchase intentions towards DS in Malaysia.

Telemedicine in the Era of Covid-19: Teleconsultation Architecture Platform

Teleconsultation is a type of medical practice similar to face-to-face consultations, and it allows a health professional to give a consultation remotely through information and communication technologies. In the context of the management of the coronavirus epidemic, the use of teleconsultation practices can facilitate healthcare access and limit the risk of avoidable propagation in medical cabinets. This paper presents the monitoring of international teleconsultation referrals in the era of Covid-19 to facilitate and prevent the suspension of access to care, the most common architecture for teleconsultation, communication technologies and protocols, vital body signals, video transmission, and the conduct of teleconsultation. The aim is to develop a teleconsultation platform to diagnose the patient in real time, transmit data from the remote location to the doctor, and provide a teleconsultation. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Mediating and moderating effects of task interdependence and creative role identity behind innovation for service: evidence from China and Australia

PurposeInnovation for service contributes to service quality and customer satisfaction, and further benefits service-centered organizations to sustain competitive advantages. However, concurrent mediating and moderating mechanisms underlying innovation for service at both the group and individual levels have been scarcely investigated. The purpose of this study is to explore multilevel mediating and moderating mechanisms behind the relationship between dual-level transformational leadership (TFL) and innovation for service at the group and individual levels.Design/methodology/approachData were collected from two countries (i.e. China and Australia). Multilevel structural equation modeling was employed to validate the research model. Bootstrapping with 5,000 replications and latent moderated structural equation modeling were used to respectively examine the mediating and moderating mechanisms.FindingsThe cross-national results showed that task interdependence and creative role identity respectively played as the group-level and individual-level mediating roles between TFL and innovation for service. It was also found that task interdependence played as a cross-level predictor enhancing individual innovation for service. Task interdependence was a moderator on the relationship between individual-level TFL and creative role identity among Australian employees, but not among Chinese employees. The relationship between creative role identity and individual innovation for service was not moderated by task interdependence among both Chinese and Australian employees.Originality/valueThis study contributes to advancing the TFL–innovation research through revealing dual-level TFL as the antecedent of innovation for service at both the group and individual levels. It also extends the understandings of the mediating and moderating mechanisms behind this dual-level relationship between TFL and innovation for service.

Writing Well/Writing to Be Well: Rethinking the Purposes of Postsecondary Writing Instruction

In this article, we identify student well-being as a primary goal of postsecondary writing instruction. Reconceiving the purpose of writing instruction in this way rests on valuing the experience of writing and its impact on the writer's sense of self. We draw on posthumanist theory and empirical research to illuminate this impact as the foundation of a pedagogy that promotes students' well-being and also supports their development as writers and as human beings.

China's education response to COVID-19: A perspective of policy analysis

This study explores how China's education responses to COVID-19 from a perspective of policy analysis. Specifically, it involves building an educational policy system for COVID-19 to examine educational governance framework, school management and teaching, policies for teachers during the epidemic. The education policy during the epidemic has achieved positive results. Those results aim to ensure the physical and mental health of teachers and students, ensure the supply of epidemic prevention materials and educational resources, ensure the quality of students' learning, and enhance the application ability of teachers' teaching technology. The process of policy changes and effects of policy implementation have been examined to analyze how China's education responses to COVID-19. The characteristics and experience of China's education policy in response to the epidemic concentrate on forming a governance system under the centralized and unified leadership of the Communist Party of China, building a pattern in which families and schools cooperate closely to promote the smooth development of education and teaching, transforming from an emergency substitute during the crisis into an important motivation to promote the transformation of education paradigm, and paying great attention to remote and poor areas and disadvantaged student groups.

Stronger than partisanship and motivated reasoning: news exposure and news frames predicting US state-level preventive behaviours against COVID-19

The COVID-19 pandemic has become a partisan issue rather than an independent public health issue in the US. This study examined the behavioural consequences of motivated reasoning and framing by investigating the impacts of COVID-19 news exposure and news frames, as apparent through a Latent Dirichlet topic modelling analysis of local news coverage, on state-level preventive behaviours as understood through a nationally representative survey. Findings suggested that the media effects on various preventive behaviours differed. The overall exposure rate to all COVID-19 news articles increased mask-wearing but did not significantly impact other preventive behaviours. Four news frames significantly increased avoiding contact or avoiding public or crowded places. However, news articles discussing anxiety and stay at home order triggered resistance and countereffects and led to risky behaviours. ‘Solid Republican' state residents were less likely to avoid contact, avoid public or crowded places, and wear masks. However, partisan leanings did not interfere with the impact of differing local COVID-19 news frames on reported preventive behaviours. Plus, statements regarding pre-existing trust in Trump did not correlate with reported preventive behaviour. Attention to effect sizes revealed that news exposure and news frames could have a bigger impact on health behaviours than motivated reasoning.

IDDF2023-ABS-0032 Active inflammatory bowel disease is associated with severe covid-19: a systematic review and meta-analysis

IDDF2023-ABS-0032 Figure 1 IDDF2023-ABS-0032 Figure 2 IDDF2023-ABS-0032 Figure 3 IDDF2023-ABS-0032 Figure 4COVID-19 outcomes in moderate-severe vs mild or quiescent IBD[Figure omitted. See PDF]ConclusionsPatients with IBD, particularly UC had an increased risk of developing severe COVID-19. Active IBD is associated with adverse COVID-19 outcomes, and the risk is increased with the disease activity of IBD.

Exploring Influential Factors of Free-Floating Bike-Sharing Usage Frequency before and after COVID-19

In order to better understand the impact of COVID-19 on the free-floating bike-sharing (FFBS) system and the potential role of FFBS played in the pandemic period, this study explores the impact mechanism of travel frequency of FFBS users before and after the pandemic. Using the online questionnaire collected in Nanjing, China, we first analyze the changes of travel frequency, travel distance, and travel duration in these two periods. Then, two ordered logit models are applied to explore the contributing factors of the weekly trip frequency of FFBS users before and after COVID-19. The results show that: (1) While the overall travel duration and travel distance of FFBS users decreased after the pandemic, the trip frequency of FFBS users increased as the travel duration increased. (2) Since COVID-19, attitude perception variables of the comfort level and the low travel price have had significantly positive impacts on the weekly trip frequency of FFBS users. (3) Respondents who use FFBS as a substitution for public transport are more likely to travel frequently in a week after the outbreak of COVID-19. (4) The travel time in off-peak hours of working days, weekends, and holidays has a significantly positive correlation with the trip frequency of FFBS users. Finally, several relevant policy recommendations and management strategies are proposed for the operation and development of FFBS during the similar disruptive public health crisis.

SAUNet++: an automatic segmentation model of COVID-19 lesion from CT slices

The coronavirus disease 2019 (COVID-19) epidemic has spread worldwide and the healthcare system is in crisis. Accurate, automated and rapid segmentation of COVID-19 lesion in computed tomography (CT) images can help doctors diagnose and provide prognostic information. However, the variety of lesions and small regions of early lesion complicate their segmentation. To solve these problems, we propose a new SAUNet++ model with squeeze excitation residual (SER) module and atrous spatial pyramid pooling (ASPP) module. The SER module can assign more weights to more important channels and mitigate the problem of gradient disappearance;the ASPP module can obtain context information by atrous convolution using various sampling rates. In addition, the generalized dice loss (GDL) can reduce the correlation between lesion size and dice loss, and is introduced to solve the problem of small regions segmentation of COVID-19 lesion. We collected multinational CT scan data from China, Italy and Russia and conducted extensive comparative and ablation studies. The experimental results demonstrated that our method outperforms state-of-the-art models and can effectively improve the accuracy of COVID-19 lesion segmentation on the dice similarity coefficient (our: 87.38% vs. U-Net++: 84.25%), sensitivity (our: 93.28% vs. U-Net++: 89.85%) and Hausdorff distance (our: 19.99 mm vs. U-Net++: 26.79 mm), respectively.

Concept Drift Adaptation Methods under the Deep Learning Framework: A Literature Review

With the advent of the fourth industrial revolution, data-driven decision making has also become an integral part of decision making. At the same time, deep learning is one of the core technologies of the fourth industrial revolution that have become vital in decision making. However, in the era of epidemics and big data, the volume of data has increased dramatically while the sources have become progressively more complex, making data distribution highly susceptible to change. These situations can easily lead to concept drift, which directly affects the effectiveness of prediction models. How to cope with such complex situations and make timely and accurate decisions from multiple perspectives is a challenging research issue. To address this challenge, we summarize concept drift adaptation methods under the deep learning framework, which is beneficial to help decision makers make better decisions and analyze the causes of concept drift. First, we provide an overall introduction to concept drift, including the definition, causes, types, and process of concept drift adaptation methods under the deep learning framework. Second, we summarize concept drift adaptation methods in terms of discriminative learning, generative learning, hybrid learning, and others. For each aspect, we elaborate on the update modes, detection modes, and adaptation drift types of concept drift adaptation methods. In addition, we briefly describe the characteristics and application fields of deep learning algorithms using concept drift adaptation methods. Finally, we summarize common datasets and evaluation metrics and present future directions.