Masked Face Recognition System using Extended VGG-19

The scientific community has observed several issues as a result of COVID-19, both directly and indirectly. The use of face mask for health protection is crucial in the current COVID-19 scenario. Besides, ensuring the security of all people, from individuals to the state system, financial resources, diverse establishments, government, and non-government entities, is an essential component of contemporary life. Face recognition system is one of the most widely used security technology in modern life. In the presence of face masks, the performance of the current face recognition systems is not satisfactory. In this paper, we investigate a flexible solution that could be employed to recognize masked faces effectively. To do this, we develop a unique dataset to recognize the masked face, consisting of a frontal and lateral face with a mask. We propose an extended VGG19 deep model to improve the accuracy of the masked face recognition system. Then, we compare the accuracy of the proposed framework to that of well-known deep learning techniques, such as the standard Convolutional Neural Network (CNN) and the original VGG19. The experimental results demonstrate that the proposed extended VGG19 outperforms the investigated approaches. Quantitatively, the proposed model recognizes the frontal face with the mask with high accuracy of 96%. © 2022 IEEE.

The role of organisational commitment and leader-member exchange in knowledge application during the COVID-19 pandemic

Purpose: This study aims to examine the role of organisational commitment (affective, normative, continuance) in influencing employees’ knowledge application behaviour during the COVID-19 pandemic. This study also probes the moderating role of leader–member exchange (LMX) in the association between organisational commitment and knowledge application. Design/methodology/approach: This study used a sample of 206 employees working in various private sector organisations in Brunei Darussalam. Structural equation modelling using Smart-PLS was used to test the hypothesised relationships. Findings: The findings show that affective and normative organisational commitment spurred employees’ knowledge application behaviour significantly during the COVID-19 crisis. However, the moderating effect of LMX could not be established in this study. Practical implications: The findings provide managers with insights into the crucial role organisational commitment can play in encouraging knowledge application in an organisation. Originality/value: Studies exploring the enabling factors of knowledge application are scarce, especially in the context of a global crisis such as the COVID-19 pandemic. This study develops a model and empirically validates the importance of organisational commitment for knowledge application amidst the COVID-19 pandemic. This study also provides insights for managers into how LMX can affect knowledge application outcomes, particularly during uncertain times. © 2022, Emerald Publishing Limited.

Improved Transfer Learning Architecture to Classify Covid-19 Affected Chest X-Rays using Noisy Student Pre-training

During this pandemic situation Chest, X-rays may play a vital role in the diagnosis of COVID-19. The shortage of labeled medical images becomes this diagnosis more challenging. We established an efficient transfer learning method for classifying COVID-19 chest X-rays. We also gathered images from the publicly available chest x-ray datasets. We built an effective classifier for our pre-trained model with the latest state-of-the-art activation function Mish, Batch Normalization, and Dropout Layer. Our classifier efficiently detects Covid-19, Pneumonia, and normal case by differentiating inflammation in the lungs. Furthermore, we used the recent state-of-the-art idea of semi-supervised Noisy Student Training in our EfficientNet Architecture model and compared it with other benchmark models. We found that our proposed model performs well by using benchmark evaluation metrics(accuracy, F1 score, and ROC(AUC)) and our ROC(AUC) score of 98% overall. After that, we visually interpreted our training model with a saliency map to make it more understandable. Contribution: We contributed an improved three-class classifier part using the new state-of-the-art activation function Mish for the EfficientNet Transfer Learning model and improved the accuracy of Covid-19 Classification through Semi-Supervised Noisy Student training. © 2021 IEEE

Combating Domestic Violence during COVID-19 Pandemic in Bangladesh: Using a Mobile Application integrated with an Effective Solution

Domestic violence (DV) is not new. Yet, researchers and human rights experts are reporting an alarming rise in DV against women since countries began locking down areas to stop the virus from spreading. They are now calling for a way to assist victims without risking infection of the virus. This paper proposes a mobile application that addresses DV in the context of the COVID-19 pandemic. It consists of two interconnected components-the Women's Support Division (WSD) and Conversational Interactive Response (CIR). The CIR module is a Natural Language Processing (NLP) chatbot for answering questions. One of the sub-modules, the Action Key module sends notifications with GPS location to acquaintances. It was found to have a performance time of 8.64 milliseconds which ensures the proposed system transmits data faster than other systems. The system usability scale (SUS) result of our proposed application has an average 66.71% score which indicates the system is fit for use. Finally, data for DV in Bangladesh is analyzed from 2014 to April 2020. Most applications addressed violence against women who are outdoors. Few have focused on DV, which is increasing indoors during the COVID-19 pandemic.

Risk factors associated with morbidity and mortality outcomes of COVID-19 patients on the 28th day of the disease course: a retrospective cohort study in Bangladesh.

Diverse risk factors intercede the outcomes of coronavirus disease 2019 (COVID-19). We conducted this retrospective cohort study with a cohort of 1016 COVID-19 patients diagnosed in May 2020 to identify the risk factors associated with morbidity and mortality outcomes. Data were collected by telephone-interview and reviewing records using a questionnaire and checklist. The study identified morbidity and mortality risk factors on the 28th day of the disease course. The majority of the patients were male (64.1%) and belonged to the age group 25-39 years (39.4%). Urban patients were higher in proportion than rural (69.3% vs. 30.7%). Major comorbidities included 35.0% diabetes mellitus (DM), 28.4% hypertension (HTN), 16.6% chronic obstructive pulmonary disease (COPD), and 7.8% coronary heart disease (CHD). The morbidity rate (not-cured) was 6.0%, and the mortality rate (non-survivor) was 2.5%. Morbidity risk factors included elderly (AOR = 2.56, 95% CI = 1.31-4.99), having comorbidity (AOR = 1.43, 95% CI = 0.83-2.47), and smokeless tobacco use (AOR = 2.17, 95% CI = 0.84-5.61). The morbidity risk was higher with COPD (RR = 2.68), chronic kidney disease (CKD) (RR = 3.33) and chronic liver disease (CLD) (RR = 3.99). Mortality risk factors included elderly (AOR = 7.56, 95% CI = 3.19-17.92), having comorbidity (AOR = 5.27, 95% CI = 1.88-14.79) and SLT use (AOR = 1.93, 95% CI = 0.50-7.46). The mortality risk was higher with COPD (RR = 7.30), DM (RR = 2.63), CHD (RR = 4.65), HTN (RR = 3.38), CKD (RR = 9.03), CLD (RR = 10.52) and malignant diseases (RR = 9.73). We must espouse programme interventions considering the morbidity and mortality risk factors to condense the aggressive outcomes of COVID-19.