Large-Scale Analysis of New Employee Network Dynamics

The COVID-19 pandemic has accelerated digital transformations across industries, but also introduced new challenges into workplaces, including the difficulties of effectively socializing with colleagues when working remotely. This challenge is exacerbated for new employees who need to develop workplace networks from the outset. In this paper, by analyzing a large-scale telemetry dataset of more than 10,000 Microsoft employees who joined the company in the first three months of 2022, we describe how new employees interact and telecommute with their colleagues during their "onboarding"period. Our results reveal that although new hires are gradually expanding networks over time, there still exists significant gaps between their network statistics and those of tenured employees even after the six-month onboarding phase. We also observe that heterogeneity exists among new employees in how their networks change over time, where employees whose job tasks do not necessarily require extensive and diverse connections could be at a disadvantaged position in this onboarding process. By investigating how web-based people recommendations in organizational knowledge base facilitate new employees naturally expand their networks, we also demonstrate the potential of web-based applications for addressing the aforementioned socialization challenges. Altogether, our findings provide insights on new employee network dynamics in remote and hybrid work environments, which may help guide organizational leaders and web application developers on quantifying and improving the socialization experiences of new employees in digital workplaces. © 2023 ACM.

COVID-19 Image Segmentation Algorithms Based on Conditional Generative Adversarial Network

The size and location of the lesions in CT images of novel corona virus pneumonia (COVID-19) change all the time, and the lesion areas have low contrast and blurred boundaries, resulting in difficult segmentation. To solve this problem, a COVID-19 image segmentation algorithm based on conditional generative adversarial network (CGAN) is proposed. Uses the improved DeeplabV3+ network as a generator, which enhances the extraction of multi-scale contextual features, reduces the number of network parameters and improves the training speed. A Markov discriminator with 6 fully convolutional layers is proposed instead of a common discriminator, with the aim of focusing more on the local features of the CT image. By continuously adversarial training between the generator and the discriminator, the network weights are optimised so that the final segmented image generated by the generator is infinitely close to the ground truth. On the COVID-19 CT public dataset, the area under the curve of ROC, F1-Score and dice similarity coefficient achieved 96.64%, 84.15% and 86.14% respectively. The experimental results show that the proposed algorithm is accurate and robust, and it has the possibility of becoming a safe, inexpensive, and time-saving medical assistant tool in clinical diagnosis, which provides a reference for computer-aided diagnosis. © 2022 ACM.

Multi2Claim: Generating Scientific Claims from Multi-Choice Questions for Scientific Fact-Checking

In automated scientific fact-checking, machine learning models are trained to verify scientific claims given evidence. A major bottleneck of this task is the availability of large-scale training datasets on different domains, due to the required domain expertise for data annotation. However, multiple-choice question-answering datasets are readily available across many different domains, thanks to the modern online education and assessment systems. As one of the first steps towards addressing the fact-checking dataset scarcity problem in scientific domains, we propose a pipeline for automatically converting multiple-choice questions into fact-checking data, which we call Multi2Claim. By applying the proposed pipeline, we generated two large-scale datasets for scientific-fact-checking: Med-Fact and Gsci-Fact for the medical and general science domains, respectively. These two datasets are among the first examples of large-scale scientific-fact-checking datasets. We developed baseline models for the verdict prediction task using each dataset. Additionally, we demonstrated that the datasets could be used to improve performance measured by weighted F1 on existing fact-checking datasets such as SciFact, HEALTHVER, COVID-Fact, and CLIMATE-FEVER. In some cases, the improvement in performance was up to a 26% increase. The generated datasets are publicly available. © 2023 Association for Computational Linguistics.

Monitoring, Tracking and Fighting Pandemics using Drone-based Artificial Intelligence in IoT

The world has altered since the World Health Organization (WHO) designated (COVID-19) a worldwide epidemic. Everything in society, from professions to routines, has shifted to accommodate the new reality. The World Health Organization warns that future pandemics of infectious diseases are likely and that people should be ready for the worst. Therefore, this study presents a framework for tracking and monitoring COVID-19 using a Deep Learning (DL) perfect. The suggested framework utilises UAVs (such as a quadcopter or drone) equipped with artificial intelligence (AI) and the Internet of Things (IoT) to keep an eye on and combat the spread of COVID-19. AI/IoT for COVID-19 nursing and a drone-based IoT scheme for sterilisation make up the bulk of the infrastructure. The proposed solution is based on the use of a current camera installed in a face-shield or helmet for use in emergency situations like pandemics. The developed AI algorithm processes the thermal images that have been detected using multi-scale similar convolution blocks (MPCs) and Res blocks that are trained using residual learning. When infected cases are detected, the helmet's embedded Internet of Things system can trigger the drone system to intervene. The infected population is eradicated with the help of the drone's sterilisation process. The developed system undergoes experimental evaluation, and the findings are presented. The developed outline delivers a novel and well-organized arrangement for monitoring and combating COVID-19 and additional future epidemics, as evidenced by the results. © 2023 IEEE.

Transthoracic echocardiography outcomes in critical COVID-19 and association with symptom burden - a longitudinal cohort study

Objective: To look for any relationship between severe/critical coronavirus disease 2019 (COVID-19) illness and post-discharge cardiac function, and also assess any correlation between this and post-COVID symptom burden. Design: Observational cohort study with both retrospective and prospective components. Setting: Intensive Care Unit (ICU) and subsequent outpatient clinic at a tertiary hospital in Western Sydney, New South Wales (NSW), Australia. Patients: All patients admitted to the ICU with COVID-19 infection between 01 July 2021 and 31 December 2021 were included (n=89). Interventions: The cohort was divided into survivors (n=61) and non-survivors (n=28). Those who underwent transthoracic echocardiography (TTE) (survivors, n=22;and non-survivors, n=23). The survivors who had an inpatient TTE were invited back for a repeat TTE and standardised symptom assessment questionnaire (COVID-19 Yorkshire Rehabilitation Scale [C19-YRS]). For all patients, demographic, clinical, biochemical, and pharmacologic data was collected. Measurements and results: Eighty-nine patients were included in the initial dataset, of which 45 had a TTE whilst acutely unwell, and 22/45 survived to hospital discharge. There were no significant differences in the measured TTE parameters between survivors and non-survivors. Of the survivors with a follow-up TTE, the majority of the changes seen in the initial study had resolved. Despite this, there was still an appreciable symptom burden in the domains of fatigue, breathlessness, ability to independently do activities of daily living, and overall reduced perception of health. Conclusions: In a cohort of critically unwell COVID-19 patients, there were no significant echocardiographic differences between survivors and non-survivors. For the survivors, whilst the majority of acute cardiac changes associated with COVID-19 infection resolved over time, however, there remained a significant symptom burden, including breathlessness and fatigability, suggesting a non-cardiac aetiology of these symptoms.

Learning Topical Structured Interfaces from Medical Research Literature

Accessing large-scale structured datasets such as WDC or CORD-191 is very challenging. Even if one topic (e.g. COVID-19 vaccine efficacy) is of interest, all topical tables in different sources/papers have hundreds of different schemas, depending on the authors, which significantly complicates both finding and querying them. Here we demonstrate a scalable Meta-profiler system, capable of constructing a structured standardized interface to a topic of interest in large-scale (semi-)structured datasets. This interface, that we call Meta-profile represents a multi-dimensional meta-data summary for a selected topic of interest, accumulating all differently structured representations of the topical tables in the dataset. Such Meta-profiles can be used as a rich visualization as well as a robust structural query interface simplifying access to large-scale (semi-)structured data for different user segments, such as data scientists and end users. © 2023 Owner/Author.

Beyond the Exscalate4CoV Project: LIGATE and REMEDI4ALL Projects

In the last 2 years, the SARS-CoV-2 (COVID-19) pandemic demonstrated that rapid response to outbreaks with readily effective treatments represents a primary health and societal priority. At the same time, we became conscious that technological resources are often not used in the most efficient manner. The LIGATE and REpurposing MEDIcines For All (REMEDI4ALL) projects started on the large-scale mobilization efforts of the EXaSCale smArt pLatform Against paThogEns (Exscalate4Cov) project with the aim to apply cutting-edge technologies in drug discovery, sustain the fight against future pandemics, and promote the everyday fight against rare diseases. In particular, the LIGATE project, using the drug-discovery platform Exscalate, intends to boost the virtual screening of drug campaigns at an extreme scale in terms of performance and streamline the drug-development process. The aim of the REMEDI4ALL project is to collect sciQ1entific expertise and innovative technology platforms for the repurposing of medicines to treat rare diseases or other pathologic conditions with no current therapy. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Reliability and acceptability of web-based administration of spanish ages and stages questionnaires third edition

to analyse the reliability of 6 age intervals of Spanish web-based Ages and Stages Questionnaires Third Edition (WB-ASQ-3) and parents' acceptability, in a clinical practice context, during COVID-19 pandemic and lockdown in Chile. Cross-sectional study with a longitudinal component. Internal consistency was analysed by Cronbach's alpha and Pearson product-moment correlation, Pearson's correlations between intervals questionnaires were analysed. Acceptability was determined using a parental satisfaction survey. 1061 children completed 2318 validated questionnaires. Scores were compared with normative sample, except for Gross motor at 8 and 12 months, and Communication at 12, 18 and 24 months, without differences to the previous Chileans sample. In domain analysis, Cronbach's alpha ranged from 0.36 to 0.80 and reached 0.74 to 0.81 for the overall test. Strong positive correlation between the domain and overall score was obtained. Correlation between most questionnaires at different ages was positive and significant. Parental surveys show high satisfaction;parents were thankful for being able to monitor children's development. WB-ASQ-3 is a reliable and feasible screening system, particularly when the test is taken as a whole, rather than by domains. It is remarkably high acceptability among parents, especially in complex times such as the lockdown and the COVID-19 pandemic. (PsycInfo Database Record (c) 2023 APA, all rights reserved) AD -, Paula Schonhaut, Luisa: lschonhaut@alemana.cl Schonhaut, Luisa: Clinica Alemana, Faculty of Medicine, Universidad del Desarrollo, Santiago, Chile Armijo, Ivan: Faculty of Psychology, Universidad del Desarrollo, Santiago, Chile Rojas B., Paula: Clinica Alemana, Faculty of Medicine, Universidad del Desarrollo, Santiago, Chile Cabrera, Leonardo: Clinica Alemana, Faculty of Medicine, Universidad del Desarrollo, Santiago, Chile Boisier, Roberto: Clinica Alemana, Faculty of Medicine, Universidad del Desarrollo, Santiago, Chile

TRACE-Omicron: Policy Counterfactuals to Inform Mitigation of COVID-19 Spread in the United States

The Omicron wave is the largest wave of COVID-19 pandemic to date, more than doubling any other in terms of cases and hospitalizations in the United States. In this paper, a large-scale agent-based model of policy interventions that could have been implemented to mitigate the Omicron wave is presented. The model takes into account the behaviors of individuals and their interactions with one another within a nationally representative population, as well as the efficacy of various interventions such as social distancing, mask wearing, testing, tracing, and vaccination. We use the model to simulate the impact of different policy scenarios and evaluate their potential effectiveness in controlling the spread of the virus. The results suggest the Omicron wave could have been substantially curtailed via a combination of interventions comparable in effectiveness to extreme and unpopular singular measures such as widespread closure of schools and workplaces, and highlight the importance of early and decisive action. © 2023 Wiley-VCH GmbH.

ACHIEVING EMPLOYMENT DIVIDEND IN THE POST-COVID-19 ERA: AN EXPLORATION FROM CHINA'S CARBON MARKET

Under the pressure of economic uncertainty and environmental protection in the post-COVID-19 era, achieving a new round of employment dividends has become one practical choice. Using the panel data of 30 Chinese provinces from 2007 to 2019, this study estimates the employment outcomes of carbon ETS pilots based on the difference-in-differences model. The findings of this study indicate the following: (1) Carbon ETS pilots can positively increase employment scales with an average effect of 7.12%. (2) This promoting effect will become more significant in provinces with high education levels, provinces with high average wages, and eastern region provinces. But there is no obvious difference between gender. (3) This positive effect can be transferred and enhanced by market competition and energy consumption. At the crossroads of green economic recovery, it will be greatly beneficial to formulate the national carbon market development roadmap under the carbon neutrality strategy.

COVID‐related stressors before and after delivery linked to adverse effects on mother and baby

Researchers have found that women who deliver babies in the midst of a pandemic are more likely to have mental health consequences, including postpartum depression, and their infants are likely to suffer as well, if the stressors are not mitigated. The cohort study of 318 mothers in the United States, Australia, and the United Kingdom found that COVID‐related stress was significantly associated with poor postpartum mental health and increased negative affectivity among infants.

Brain-derived neurotrophic factor (BDNF) and the capute scales in offspring of mothers with normal and deficient vitamin D levels

Background: Vitamin D deficiency during pregnancy is associated with a child's neurocognitive development. During the COVID-19 pandemic, the Capute Scales have become ineffective because requiring face-to-face interaction to examine children's neurocognitive development. Brain-Derived Neurotrophic Factor (BDNF) is an alternative method to determine children's neurocognitive development through blood sampling. This study aims to evaluate the relationships between BDNF and the Capute Scales in the offspring of mothers with vitamin D levels. Methods: A diagnostic study from 14 community health centers in Semarang City, Central Java, Indonesia, was conducted. The study was performed from August 2017 until August 2018 and included vitamin D level records at 20–24 weeks of normal gestation in 2017 from all single births at these community health centers. The cases were divided into two groups: 30 children born to mothers with normal vitamin D levels (25(OH)D > 20 ng/mL) and 30 children born to mothers with vitamin D deficiency (25(OH)D < 20 ng/ml). This study aimed to assess the relationship between BDNF levels and the Capute Scales of children aged 2 years and to determine the cut-off value based on each group's examination of BDNF levels. Results: The mean age of pregnant women whose vitamin D levels were checked was 29.32 years. The cut-off value of BDNF among the two groups was 7968 pg/mL. Of the 55 samples that met the Capute Scales criteria, 44 had a BDNF value less than the cut-off value, and 11 samples had a BDNF value more than the cut-off value. Meanwhile, of the five samples that met the Capute Scales criteria and were suspected of mental retardation, three had a BDNF value less than the cut-off value, and two had a BDNF value more than the cut-off value. The sensitivity, specificity, positive, and negative predictive values were 80%, 60%, 0.93, and 0.15, respectively. Conclusions: BDNF levels can be an alternative neurocognitive examination in children born to mothers with normal and deficient vitamin D levels. BDNF is a growth factor that plays an important role in neurons' differentiation, growth, and survival in the postnatal phase. © 2023, Sanglah General Hospital. All rights reserved.

Work-Related Musculoskeletal Disorders and Mental Health among Nursing Personnel in the Context of COVID-19 Pandemic in West Bengal, India

Introduction: COVID-19 outbreak created enormous turmoil all over the world. The health sectors were overburdened with critically ill COVID-19 patients. The load on the health sector burdened and fatigued the health care workers especially the nursing personnel as they were in direct contact with the symptomatic patients predominantly. In each wave, the patients' number surged and overwhelmed the health sectors. In this study, the prevalence of work-related musculoskeletal disorders (WMSDs) discomfort and mental exhaustion among the nursing personnel were assessed. Methods: The duration of the study was from the end of September 2021 to March 2022. The factors that sourced the discomfort symptoms amidst and the interludes of the later waves of COVID-19 and their interrelations were also assessed here. The Modified Nordic Questionnaire and the COVID Stress Scales (CSS) were used in this study. Results: 73.9% of subjects (N=88) reported the presence of WMSDs symptoms in at least one body region in the last one-year time period with the most prevalent discomfort in the lower back region (42%), followed by discomforts in the ankles, neck, and knee. The mean value of the total score of CSS was 33.74(±19.90) with high mean values in compulsive checking, danger, and contamination subscales. This study showed significant positive correlations between the presence of WMSD symptoms with BMI (rs=0.266, p= 0.012) and CSS total scores (rs=0.216, p=0.043). Conclusion: Large percentages of reports on musculoskeletal discomforts with reports on mental enervation related to COVID-19 indicated weariness of the nursing personnel. The physical exhaustion due to overloading work and the interplay between mental stress and WMSDs revealed pivotal issues of concern. © 2023 The Author(s).

[Vaccine hesitancy: current status, associated factors, measurement, and approach].

Definition and present status　Vaccine hesitancy, defined as "delay in acceptance or refusal of vaccination despite availability of vaccination services," is a global public health concern. Specifically, COVID-19 and human papillomavirus (HPV) vaccine hesitancy remains a major social challenge in Japan, and effective preventive strategies are urgently required. In this review, we discuss previous studies that have described vaccine hesitancy.Associated factors　Vaccine hesitancy is affected by several factors, primarily psychological variables (referred to as the "3Cs" comprising confidence, complacency, and convenience regarding individuals' perceptions of vaccination) and sociodemographic variables (age, sex, socioeconomic status, race, and social capital). "Behavioral and Social Drivers of Vaccination Framework", developed recently by the World Health Organization, has focused on vaccination-specific beliefs and reports that programs may affect and are likely to have wide applicability in the development of effective interventions.Measurement　Identification of psychological factors associated with vaccination hesitancy is important to establish strategies to increase vaccine uptake. Many scales are available to measure vaccine hesitancy and psychological factors that affect vaccine hesitancy. These scales include different evaluation items, validity, reliability, and availability of validated Japanese versions. Therefore, careful selection of scales based on their intended purpose and the target population in whom the desired intervention is intended are important. A representative 7C scale is widely used globally. It has been translated into more than 10 languages, including Japanese.Approach　Several studies and articles, mainly developed for the European and American populations provide guidelines for selection of evidence-based strategies and interventions to increase vaccine uptake. Evidence-based strategies may be broadly classified into the following categories: (1) Strengthening the healthcare system through implementation of the principles of behavioral science. (2) Development of tailored approaches using systematic listening activities. (3) Provision of evidence-based resources to support healthcare personnel. (4) Utilization of media. Based on findings described by previous studies discussed in this report, it may be important to plan strategies to improve the uptake of each vaccine in Japan, such as those for COVID-19, HPV, and also childhood vaccines.

Adapting the COVID Stress Scale (CSS) to investigate the level of stress among Hong Kong Chinese people 1 year after the COVID-19 pandemic.

BACKGROUND: Stress is a crucial driver that affects hygiene behavior. The Hong Kong population lacks a COVID-19 or pandemic related stress measure investigating the COVID-19 related stress after one year of outbreak. DESIGN AND METHODS: The original COVID Stress Scale (CSS) was translated and culturally adapted into the Chinese (Cantonese) version (CSS-C). Six hundred and twenty-four participants were recruited from the general public to examine the internal consistency, and concurrent and convergent validity of the CSS-C. The test-retest reliability of CSS-C was examined using 39 university students. RESULTS: People with old age, women, single, low educational level and borderline and abnormal levels of anxiety and depression were likely to perceive high level of COVID-19 related stress. All CSS-C subscales demonstrated good internal consistency, moderate to good test-retest reliability, and weak to moderate correlations with various mental health-related measures. DISCUSSION: The CSS could help monitor the stress associated the current and potential future pandemics.

First in-neonate use of non-invasive transcutaneous auricular vagus nerve stimulation: 18-month neurodevelopmental and sensory follow-up

Objective: Non-invasive transcutaneous auricular vagal nerve stimulation (taVNS) paired with oral feeding is a novel intervention for infants with feeding delays that may improve feeding and help avoid a gastrostomy tube (Gtube). However, the long-term impact of this neurostimulation on infant's development remains unknown. We investigated the neurodevelopmental and sensory outcomes of infants who received taVNS paired with bottle feeding. Method(s): Twenty-one of 35 toddlers who participated in the open label trial of taVNS paired with one or two feeds a day for 2-3 weeks, underwent comprehensive developmental assessments at 18 months of age using Cognitive Adaptive Test, Clinical Linguistics and Auditory Milestone, and Peabody gross motor scores. Twelve of those assessed achieved full oral feeds ('responders') and 9 had G-tube placed for feeds ('non-responders'). Before COVID, 12 toddlers (5 responders, 7 non-responders) were also assessed in the home using the Bayley-III and Sensory Profile (SP-2) assessments. Area deprivation index (ADI) was used to measure resource poor environments and relate to test scores. We used Fishers exact test and Pearson correlation coefficients to compare neurodevelopmental and sensory performance in responders versus non-responders. Result(s): taVNS responders showed significantly better general sensory processing in SP-2 than did non-responders (p =0.04). There were no significant differences in Bayley-III or CAT/CLAMS/ASQ scores in areas of cognition, receptive language, fine motor, and gross motor skills in this small sample size, but are similar to published scores for preterm infants who received G-tubes. ADI was not significantly associated with neurodevelopmental scores. Conclusion(s): These results suggest that taVNS paired with feeding may have a potential long-term positive neurodevelopmental effect on sensory processing in neonates with poor feeding. The current open-label results need testing in randomized controlled trials of taVNS paired with oral feeding in developmentally delayed infants failing oral feeds. Research Category and Technology and Methods Clinical Research: 12. Vagus Nerve Stimulation (VNS) Keywords: Neurodevelopment, taVNS, feeding, developmental delaysCopyright © 2023

Scale matters: Variations in spatial and temporal patterns of epidemic outbreaks in agent-based models

Agent-based modellers frequently make use of techniques to render simulated populations more computationally tractable on actionable timescales. Many generate a relatively small number of "representative” agents, each of which is "scaled up” to represent some larger number of individuals involved in the system being studied. The degree to which this "scaling” has implications for model forecasts is an underdeveloped field of study;in particular, there has been little known research on the spatial implications of such techniques. This work presents a case study of the impact of the simulated population size, using a model of the spread of COVID-19 among districts in Zimbabwe for the underlying system being studied. The impact of the relative scale of the population is explored in conjunction with the spatial setup, and crucial model parameters are varied to highlight where scaled down populations can be safely used and where modellers should be cautious. The results imply that in particular, different geographical dynamics of the spread of disease are associated with varying population sizes, with implications for researchers seeking to use scaled populations in their research. This article is an extension on work previously presented as part of the International Conference on Computational Science 2022 (Wise et al., 2022)[1]. © 2023 The Authors

The Collages of Athens. The Representation of the City in the 21th Century

Since 2000s, Athens has been changed due to imponent urban transformations mainly provoked by the implementation of large-scale works in occasion of the 28th Olympic Games in 2004, then by the dramatic effects caused by the impact of the 2008 economic crisis and the imposition of austerity policies and, finally, by the unbalanced growth of the subsequent economic recovery that abruptly halted with the impact of the Covid-19 pandemic. The city has rapidly changed not only its urban configuration, but also the way it has been perceived and represented. During these years, it can be registered a diffuse interest among architects and artists for the use of collage to represent Athens. While during the previous century in Greece collage was mainly limited to the representation of rural and mythological landscape, since 2000s many collaged images embed for the first time many urban features of the contemporary metropolis. This paper aims to present how collage art had been adopted by architects to describe, analyze, critic, and imagine the city as an attempt to find a proper tool to deal effectively with the emerging urban issues that had arose with the new status of Athens as a metropolitan city. In particular, the association between collage and the city will be investigated in relation with the periods of growth (2000–2008), decay (2008–2015) and economic recovery (2015–2020). © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Evaluating the feasibility of delivering a sleep education programme in secondary schools

Background: School-based sleep education programmes can promote the importance of sleep health and may improve adolescent sleep. To date, only limited research has examined the feasibility of integrating sleep programmes into the school curriculum. Objectives: This study evaluated the feasibility, acceptability and preliminary efficacy of the Strathclyde intervention to encourage good sleep health in teenagers (SIESTA). Methods: A total 171 students (12–15 years, 53% women) from secondary schools in Scotland participated in the study. Recruitment and retention, data collection and design procedures were assessed to establish feasibility. Qualitative feedback on acceptability was collected via focus group discussions. Outcome measures assessing insomnia symptoms, sleep hygiene, depression, anxiety and stress were completed at baseline and post-intervention to explore the preliminary effects of SIESTA. Results: All schools that were approached consented to participate, and most students completed assessments at both time points (171) with a dropout rate of 5%. Assessment measures provided sufficient data to compare baseline and post-intervention values. Training and delivery manuals ensured successful delivery of the programme. Qualitative feedback indicated SIESTA was acceptable, and students spoke favourably about the content, delivery and techniques. Students reported that SIESTA was age-appropriate, relevant and the techniques were beneficial. There were significant improvements in insomnia and stress, but no improvements were noted for sleep hygiene, depression or anxiety. Conclusion: The findings suggest that SIESTA is feasible and acceptable for delivery via the school curriculum. The results indicate that a controlled trial is required to further investigate the efficacy of SIESTA implemented in an educational context.

Generating Masked Facial Datasets Using Dlib-Machine Learning Library

In 2020, the COVID-19 pandemic spread globally, leading to countries imposing health restrictions on people, including wearing masks, to prevent the spread of the disease. Wearing a mask significantly decreases distinguishing ability due to its concealment of the main facial features. After the outbreak of the pandemic, the existing datasets became unsuitable because they did not contain images of people wearing masks. To address the shortage of large-scale masked faces datasets, a developed method was proposed to generate artificial masks and place them on the faces in the unmasked faces dataset to generate the masked faces dataset. Following the proposed method, masked faces are generated in two steps. First, the face is detected in the unmasked image, and then the detected face image is aligned. The second step is to overlay the mask on the cropped face images using the dlib-ml library. Depending on the proposed method, two datasets of masked faces called masked-dataset-1 and masked-dataset-2 were created. Promising results were obtained when they were evaluated using the Labeled Faces in the Wild (LFW) dataset, and two of the state-of-the-art facial recognition systems for evaluation are FaceNet and ArcFace, where the accuracy of using the two systems was 96.1 and 97, respectively with masked-dataset-1 and 87.6 and 88.9, respectively with masked-dataset-2. © 2022 IEEE.