Assessing Robustness of a Deep Learning Model for COVID-19 Classification on Chest Radiographs

The COVID-19 pandemic has made a dramatic impact on human life, medical systems, and financial resources. Due to the disease's pervasive nature, many different and interdisciplinary fields of research pivoted to study the disease. For example, deep learning (DL) techniques were employed early to assess patient diagnosis and prognosis from chest radiographs (CXRs) and computed tomography (CT) scans. While the use of artificial intelligence (AI) in the medical sector has displayed promising results, DL may suffer from lack of reproducibility and generalizability. In this study, the robustness of a pre-trained DL model utilizing the DenseNet-121 architecture was evaluated by using a larger collection of CXRs from the same institution that provided the original model with its test and training datasets. The current test set contained a larger span of dates, incorporated different strains of the virus, and included different immunization statuses. Considering differences in these factors, model performance between the original and current test sets was evaluated using area under the receiver operating characteristic curve (ROC AUC) [95% CI]. Statistical comparisons were performed using the Delong, Kolmogorov-Smirnov, and Wilcoxon rank-sum tests. Uniform manifold approximation and projection (UMAP) was used to help visualize whether underlying causes were responsible for differences in performance between test sets. In the task of classifying between COVID-positive and COVID-negative patients, the DL model achieved an AUC of 0.67 [0.65, 0.70], compared with the original performance of 0.76 [0.73, 0.79]. The results of this study suggest that underlying biases or overfitting may hinder performance when generalizing the model. © 2023 SPIE.

ASSOCIATION RULE MINING FOR PREDICTION OF COVID-19

COVID-19 is a raging pandemic that has created havoc with its impact ranging from loss of millions of human lives to social and economic disruptions of the entire world. Therefore, error-free prediction, quick diagnosis, disease identification, isolation and treatment of a COVID patient have become extremely important. Nowadays, mining knowledge and providing scientific decision making for diagnosis of diseases from clinical datasets has found wide-ranging applications in healthcare sector. In this direction, among different data mining tools, association rule mining has already emerged out as a popular technique to extract invaluable information and develop important knowledge-base to help in intelligent diagnosis of distinct diseases quickly and automatically. In this paper, based on 5434 records of COVID cases collected from a popular data science community and using Rapid Miner Studio software, an attempt is put forward to develop a predictive model based on frequent pattern growth algorithm of association rule mining to determine the likelihood of COVID-19 in a patient. It identifies breathing problem, fever, dry cough, sore throat, abroad travel and attended large gathering as the main indicators of COVID-19. Employing the same clinical dataset, a linear regression model is also proposed having a moderately high coefficient of determination of 0.739 in accurately predicting the occurrence of COVID-19. A decision support system can also be developed using the association rules to ease out and automate early detection of other diseases. © 2023 by the authors.

Contact Tracing over Uncertain Indoor Positioning Data

Pandemics often cause dramatic losses of human lives and impact our societies in many aspects such as public health, tourism, and economy. To contain the spread of an epidemic like COVID-19, efficient and effective contact tracing is important, especially in indoor venues where the risk of infection is higher. In this work, we formulate and study a novel query called Indoor Contact Query (<sc>ICQ</sc>) over raw, uncertain indoor positioning data that digitalizes people's movements indoors. Given a query object <inline-formula><tex-math notation="LaTeX">$o$</tex-math></inline-formula>, e.g., a person confirmed to be a virus carrier, an <sc>ICQ</sc> analyzes uncertain indoor positioning data to find objects that most likely had close contact with <inline-formula><tex-math notation="LaTeX">$o$</tex-math></inline-formula> for a long period of time. To process <sc>ICQ</sc>, we propose a set of techniques. First, we design an enhanced indoor graph model to organize different types of data necessary for <sc>ICQ</sc>. Second, for indoor moving objects, we devise methods to determine uncertain regions and to derive positioning samples missing in the raw data. Third, we propose a query processing framework with a close contact determination method, a search algorithm, and the acceleration strategies. We conduct extensive experiments on synthetic and real datasets to evaluate our proposals. The results demonstrate the efficiency and effectiveness of our proposals. IEEE

EMPLOYING ARTIFICIAL NEURAL NETWORKS AND MULTIPLE DISCRIMINANT ANALYSIS TO EVALUATE THE IMPACT OF THE COVID-19 PANDEMIC ON THE FINANCIAL STATUS OF JORDANIAN COMPANIES

Aim/Purpose This paper aims to empirically quantify the financial distress caused by the COVID-19 pandemic on companies listed on Amman Stock Exchange (ASE). The paper also aims to identify the most important predictors of financial distress pre- and mid-pandemic. Background The COVID-19 pandemic has had a huge toll, not only on human lives but also on many businesses. This provided the impetus to assess the impact of the pandemic on the financial status of Jordanian companies. Methodology The initial sample comprised 165 companies, which was cleansed and reduced to 84 companies as per data availability. Financial data pertaining to the 84 companies were collected over a two-year period, 2019 and 2020, to empirically quantify the impact of the pandemic on companies in the dataset. Two approaches were employed. The first approach involved using Multiple Discriminant Analysis (MDA) based on Altman's (1968) model to obtain the Z-score of each company over the investigation period. The second approach involved developing models using Artificial Neural Networks (ANNs) with 15 standard financial ratios to find out the most important variables in predicting financial distress and create an accurate Financial Distress Prediction (FDP) model. Contribution This research contributes by providing a better understanding of how financial distress predictors perform during dynamic and risky times. The research confirmed that in spite of the negative impact of COVID-19 on the financial health of companies, the main predictors of financial distress remained relatively steadfast. This indicates that standard financial distress predictors can be regarded as being impervious to extraneous financial and/or health calamities. Findings Results using MDA indicated that more than 63% of companies in the dataset have a lower Z-score in 2020 when compared to 2019. There was also an 8% increase in distressed companies in 2020, and around 6% of companies came to be no longer healthy. As for the models built using ANNs, results show that the most important variable in predicting financial distress is the Return on Capital. The predictive accuracy for the 2019 and 2020 models measured using the area under the Receiver Operating Characteristic (ROC) graph was 87.5% and 97.6%, respectively. Recommendations Decision makers and top management are encouraged to focus on the identified for Practitioners highly liquid ratios to make thoughtful decisions and initiate preemptive actions to avoid organizational failure. Recommendations This research can be considered a stepping stone to investigating the impact of for Researchers COVID-19 on the financial status of companies. Researchers are recommended to replicate the methods used in this research across various business sectors to understand the financial dynamics of companies during uncertain times. Impact on Society Stakeholders in Jordanian-listed companies should concentrate on the list of most important predictors of financial distress as presented in this study. Future Research Future research may focus on expanding the scope of this study by including other geographical locations to check for the generalisability of the results. Future research may also include post-COVID-19 data to check for changes in results. © 2023 Informing Science Institute. All rights reserved.

CovRelex-SE: Adding Semantic Information for Relation Search via Sequence Embedding

In recent years, COVID-19 has impacted all aspects of human life. As a result, numerous publications relating to this disease have been issued. Due to the massive volume of publications, some retrieval systems have been developed to provide researchers with useful information. In these systems, lexical searching methods are widely used, which raises many issues related to acronyms, synonyms, and rare keywrds. In this paper, we present a hybrid relation retrieval system, CovRelex-SE, based on embeddings to provide high-quality search results. Our system can be accessed through the following URL: https://www.jaist.ac.jp/is/labs/nguyen-lab/systems/covrelex-se/. © 2023 Association for Computational Linguistics.

Close Contact Individual Monitoring Based on Prescribed Social Distance Using Computer Vision

Contact tracing is one of the methods used by the government and organizations for controlling viral diseases like COVID-19, which claimed many human lives. Social distancing is advised to everyone to minimize the virus from spreading. This study aims to build a contact tracing tool that monitors social distancing individually using computer vision in real-time. Object tracking by detection is used for individual monitoring with YOLOv4 (You Only Look Once) as the object detector and SORT (Simple Online and Real-time Tracking) as the object tracker. The combination gained an average streaming and detection frame rate of 26 FPS and 10 FPS on NVIDIA's GTX 1650, respectively. It is expected to have more frame rate when used in a more powerful device. Moreover, the system obtained 98.2% accuracy in measuring the distance between individuals. Furthermore, the performance of the QR scanner used in the study attains a 100% success rate and a 98% accuracy in allocating the QR code to the correct owner from the video stream. © 2022 IEEE.

IoE-Based Predictive Oxygen Inventory Management

The COVID-19 pandemic highlighted a major flaw in the current medical oxygen supply chain and inventory management system. This shortcoming caused the deaths of several patients which could have been avoided by accurate prediction of the oxygen demand and the distribution of oxygen cylinders. To avoid such calamities in the future, this paper proposes an Internet of Everything (IoE) based solution which forecasts the demand for oxygen with 80-85% accuracy. The predicted variable of expected patients enables the system to calculate the requirement of oxygen up to the next 30 days from the initiation of data collection. The system is scalable and if implemented on a city or district level, will help in the fair distribution of medical oxygen resources and will save human lives during extreme load on the supply chain. © 2023 IEEE.

Early Prediction of COVID-19 Infection with IoT and Machine Learning

The deadly virus COVID-19 has heavily impacted all countries and brought a dramatic loss of human life. It is an unprecedented scenario and poses an extreme challenge to the healthcare sector. The disruption to society and the economy is devastating, causing millions of people to live in poverty. Most citizens live in exceptional hardship and are exposed to the contagious virus while being vulnerable due to the inaccessibility of quality healthcare services. This study introduces ubiquitous computing as a state-of-The-Art method to mitigate the spread of COVID-19 and spare more ICU beds for those truly needed. Ubiquitous computing offers a great solution with the concept of being accessible anywhere and anytime. As COVID-19 is highly complicated and unpredictable, people infected with COVID-19 may be unaware and still live on with their life. This resulted in the spread of COVID-19 being uncontrollable. Therefore, it is essential to identify the COVID-19 infection early, not only because of the mitigation of spread but also for optimal treatment. This way, the concept of wearable sensors to collect health information and use it as an input to feed into machine learning to determine COVID-19 infection or COVID-19 status monitoring is introduced in this study. © 2023 IEEE.

Real human close contact-based analysis on COVID-19 combined interventions

COVID-19 has threatened human lives. However, the efficiency of combined interventions on COVID-19 has not been accurately analyzed. In this study, an improved SEIR model considering both real human indoor close contact behaviors and personal susceptibility to COVID-19 was established. Taking Hong Kong as an example, a quantitative efficiency assessment of combined interventions (i.e. close contact reduction, vaccination, mask-wearing, school closures, workplace closures, and body temperature screening in public places) was carried out. The results showed that the infection risk of COVID-19 of students, workers, and non-workers/students were 3.1%, 8.7%, and 13.6%, respectively. The basic reproduction number R0 was equal to 1 when the close contact reduction rate was 59.9% or the vaccination rate reached 89.5%. The results could provide scientific support for interventions on COVID-19 prevention and control. © 2022 17th International Conference on Indoor Air Quality and Climate, INDOOR AIR 2022. All rights reserved.

Predicting COVID-19 from Chest X-ray Images using a New Deep Learning Architecture

The spread of coronavirus disease in late 2019 caused huge damage to human lives and forced a chaos in health care systems around the globe. Early diagnosis of this disease can help separate patients from healthy people. Therefore, precise COVID-19 detection is necessary to prevent the spread of this virus. Many artificial intelligent technologies for example deep learning models have been applied successfully for this task by employing chest X-ray images. In this paper, we propose to classify chest X-ray images using a new end-To-end convolutional neural network model. This new model consists of six convolutional blocks. Each block consists of one convolutional layer, one ReLU layer, and one max-pooling layer. The new model was applied on a challenging imbalanced COVID19 dataset of 5000 images, divided into two classes, COVID and Non-COVID. In experiments, the input image is first resized to 256×256×3 before being fed to the model. Two metrics were used to test our new model: sensitivity and specificity. A sensitivity rate of 97% was achieved along with a specificity rate of 99.32%. These results are promising when compared to other deep learning models applied on the same dataset. © 2022 IEEE.

Internet of Medical Things Enabled by Permissioned Blockchain on Distributed Storage

The recent COVID-19 pandemic has made the world suffer ravaging costs and damage to human lives, perhaps never seen in modern world. Pandemics will keep reviving till such time the humans attain a disease-less world state. Till such realizations are attained, we need to attempt retarding the pandemics by exploiting information systems enabled with new genre IT technologies, and blockchain offers one such way for realization. This paper proposes medical IoT architecture enabled by blockchain and further augmented with distributed storage protocol to retard any such pandemics ahead. The works have been performed on a Multichain permissioned blockchain platform and IPFS protocol. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Analysing Shifts in Perceptions of Indians during COVID-19 Pandemic by Mining Tweets

Current pandemic situation has a significant impact affecting human life not only socially and economically, but emotionally and psychologically as well. This impact can be easily observed on social media platforms. Along with the knowledge exchange related to Covid-19 pandemic on social media, there is an emotional trauma wave that can be felt by carefully analyzing the activities of this social media. Keeping this view in thought, we analyze around 12000 tweets of Indian people to find out whether there is a trend shift of thinking pattern and mindset of Indian people as the pandemic progresses. The study is bifurcated into stages to clearly see the paradigm shift. We use tweets since twitter is a rich medium that can be leveraged to its optimum to have a good amount of understanding of the sentiments of the people. Analyzing the twitter dataset, we derive results and find out whether the amount of negative tweets v/s positive (or motivational) tweets have increased or not as the pandemic progresses. The study is supported by graphical visualizations of the polarity of the tweets month wise. Further, Wordmap approach is used to perform qualitative mining analysis in addition to the sentiment score based calculation. This work helps us to understand how the public opinions are changing with the changes in the spread dynamics of the virus. This kind of mood mining helps in identifying the Covid-19 situation from the psychological perspective that whether there is a sense of fear among people or they are quite optimistic of the situation. It can help in a great extend to the strategic and decision making bodies to plan out for future decisions. Further, such kind of studies can be used as reference to provide insights about mental health of people for any future incident or event of such nature. © 2023 IEEE.

Are Animal Disease Reservoirs at Risk of Human Antiviral Exposure?

Novel viral pathogens are causing diseases to emerge in humans, a challenge to which society has responded with technological innovations such as antiviral therapies. Antivirals can be rapidly deployed to mitigate severe disease, and with vaccines, they can save human lives and provide a long-term safety net against new viral diseases. Yet with these advances come unforeseen consequences when antivirals are inevitably released to the environment. Using SARS-CoV-2 as a case study, we identify global patterns of overlap between bats and elevated pharmaceutical concentrations in surface waters. We model how freshwater contamination by antivirals could result in exposure to insectivorous bats via consumption of emergent insects with aquatic larvae, ultimately risking the evolution of antiviral-resistant viruses in bats. The consequences of widespread antiviral usage for both human and ecosystem health underscore urgent frontiers in scientific research, antiviral development, and use. © 2023 American Chemical Society

Surveillance Rover to Maintain Social Distancing in Crowded Areas

COVID-19 is making a huge impact both in terms of the economy and human lives. Many lost their lives due to COVID-19 which is found in most of the nations. The number of positive symptoms is increasing rapidly all over the world. To safeguard us from the virus, some protocols have been addressed by WHO in which people has to wear a mask and make a social distancing when moved in public. Therefore, social distancing places an important role in preventing us from the spread of the diseases. The minimum distance between to be maintained is informed at 6 feet informed by the health organizations. When people gathered on a group social distancing could not be maintained even if manual or any kind of technology implemented. Temperature measurement on mass gathering was also a tedious process where the monitoring is essential. Multiple methods such as thermal cameras, temperature sensors for monitoring the personnel has not been efficient. In the proposed work to monitor the social distancing between the persons an ultrasonic sensor is placed to detect the obstacle and an IR sensor to make the rover move. An encoder is used to calculate the distance based on the rpm of the wheel. Based on this input the distance is checked within this limit the obstacle is detected, an alert signal is made using the buzzer. A thermal sensor is used to measure the temperature of the person and an LCD display shows the temperature of the person and distance between obstacles. The proposed system has resulted in identifying the distance and helps in reducing the spread during the pandemic situation. © 2023 IEEE.

Artificial Intelligence and Deep Learning based Face Mask Detection System using Generic Camera

In the past two years 2020 and 2021, the COVID-19 outburst has had a serious effect on human life. The effects and side effects of COVID-19 are being stroked in almost every discipline relevant to survival and development. The healthcare system was in a problematic situation during this tough time in pandemic situation all over the world. One of the many precautions and protections used to break the chain of spreading of this virus is wearing a mask and keeping safe distance. In a network of smart cities where entirely public spaces are monitored by Closed-Circuit Television (CCTV) cameras, we are offering a strategy in this study that restricts the spread of COVID-19 by identifying people not wearing mask. The network alerts the proper authority whenever a person without a mask is discovered. It is believed that our research will help many countries throughout the world stop the spread of this infectious disease. We've examined this on 200 real people in the present, with a 100% success rate. It is also observed that when more than one person in front of CCTV success rate reduced exponentially © 2022 IEEE.

Sentiment Analysis of Google Classroom Application Using Support Vector Machine

Technological advances are growing very rapidly and affect human life and also have an impact during the pandemic. During the pandemic, Google Classroom ranks first on the platform that is often used during distance learning. On the Google Play site, many reviews about the application. Due to the many reviews given by users, sentiment analysis is carried out so that public opinion on the application is known, thus users or application developers know the advantages and disadvantages of the application. Sentiment analysis was carried out using the SVM algorithm with the help of the python programming language on google colab. From the results of sentiment analysis, the sentiment of Google Classroom application users is 64% negative sentiment and 36% positive sentiment with an algorithm accuracy of 81%. In addition, data visualization is carried out, so that it can be seen what causes users to give positive reviews and negative reviews, from which priority improvements are obtained to improve Google Classroom performance. © 2022 IEEE.

The Impact of Covid-19 on Student Mental Health and Online Learning Experience

The ongoing coronavirus pandemic has affected every facet of human life in the contemporary world. Consequently, university students have to adjust to radically change learning environments. Moreover, the movement restrictions from the government-imposed lockdowns negatively affected students' mental health due to mental issues such as stress, frustration, and depression. The pandemic has caused considerable changes in our daily lives. These reasons are why the virus has hurt individuals' mental health, especially students who had to cope with changes in the education system and even the loss of loved ones. The ambiguity resulting from the pandemic has yet to be fully covered, particularly the students' well-being and the new learning landscape that they are anticipated to navigate seamlessly without their usual support systems. Covid-19 did disrupt the normal and put us all in numerous stressful circumstances' and forced us to have to face overwhelming difficulties at a time. Covid-19 lockdown and pandemic did bring about a sense of anxiety and fear around the world. The spectacle has led to students' long-term and short-term mental health and psychological implications. The paper presents research showing that most students were not prepared for this change, and that indeed they were affected mentally by remote learning. Additionally, the effect of prolonged pandemic fatigue and lockdown on university scholars and academic experiences is unclear. This paper reviews articles about mental health aspects of students and online learning experiences impacted by Covid-19 and provides a roadmap for an ongoing research. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Analyzing Learners' Privacy in MOOC and Online Learning Platform

COVID-19 has affected human life since its advent. And to counteract its spread, humankind adopts social distancing, which encourages remote working for employees, and online learning for students. Many universities and schools quickly adopted e-learning solutions without much consideration of security, while it is important to consider users' privacy. Unfortunately, digital learning spaces face security vulnerabilities, risks and threats and are not spared from cyber-attacks. To ensure the security and privacy of e-learning solutions used by universities and schools, we analyzed how MOOCs and Organizations offering online courses long before COVID-19 deal with their users' privacy and personal data. In this study, we considered some popular platforms from The United States (Coursera, EdX, Udemy), Europe and the United Kingdom (FutureLearn, FUN MOOC, EduOpen), and Asia (XuetangX, SWAYAM, and K-MOOC). We discussed the personal data collected by these platforms, the purposes for which these data are collected, the different legislation for processing and storing data, and how the platforms ensure user privacy. © 2022 IEEE.

Analyzing Wearable Data for Diagnosing COVID-19 Using Machine Learning Model

Health is an indispensable part of human life, but we realize its importance when we face health issues. Technology can play an important role in the healthcare sector. During the COVID-19 pandemic, many countries used technology to control the situation. Internet of Things-based wearable devices can change the whole scenario of diagnosing the disease. The physiological features collected using wearables can be used for pre-symptomatic prediction of disease. In this study, from the cohort of 185 participants, data of 36 participants are analyzed to predict COVID-19 before symptoms begin using the machine learning model. Our findings suggest that heart rate, BPM, SDNN, and steps features can be used to detect the COVID-19 before the symptoms appear. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

VIRTUAL REALITY TRAINING OF EMPLOYEES IN FOREST-BASED SMES AS A PART OF A CRISIS MANAGEMENT

Science and manufacturing have always been a generator and conduit of innovations in every field of human life. The innovations are of both fundamental and purely applied nature. The first environment for testing these innovations is the internal firm's educational system. In this regard, the last two years circumstances around the pandemic of COVID-19 served as a catalyst for the training in companies to adopt contemporary, interactive and attractive methods of training processes. Of course, some of these methods have been used in the pre-pandemic environment, but they have not been widespread. This confirms the rule related to a crisis management, namely that any crisis must be seen not only as a threat, but also as an opportunity to master new approaches and to show their effectiveness in practice. The aim of this paper is to focus on the possibilities of using virtual reality in training employees in forest-based SMEs such as specific manufacturing procedures, healthy work condition, organization of manufacturing etc. A number of research methods will be used. These will include: literature research, retrospective analysis, method of comparison etc. © 2022 15th International Scientific Conference WoodEMA 2022 - Crisis Management and Safety Foresight in Forest-Based Sector and SMES Operating in the Global Environment. All rights reserved.