Computational Intelligence Techniques for Combating COVID-19

The outbreak of pneumonia in December 2019 in Wuhan, China, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has spread rapidly throughout the world. With over 4.62 million confirmed cases of COVID-19 and 311,000 deaths in more than 188 countries, this ongoing pandemic has wreaked havoc all around the globe. However, the SAARC (South Asian Association for Regional Cooperation) countries, compared to the First World nations, have significantly low death rate. In this paper, the authors have determined this uneven distribution of COVID-19 deaths with the help of some possible factors, which are the prime cause of such variability among the different nations. This paper presents the significance of these factors through analysis of the data corresponding to each of these factors from 165 different countries. On the basis of the relationship between the factors and their significance on the concerned countries’ death toll, we have labelled each factor’s risk index using the multiple analytical hierarchy process (M-AHP), as it provides several experts’ views instead of a single expert’s opinion. The risk index of all the factors has been used to generate the susceptibility of COVID-19 for each of the countries in study, specifically the SAARC nations. Finally, we have applied a hierarchical clustering-based machine learning approach to visualize the countries’ death toll corresponding to their susceptibility index. This paper’s major findings are that authors holistically searched the root causes of why South Asian countries are less fatal concerning COVID-19.

The recently identified viral disease caused by the coronavirus is named as COVID-19. Many COVID-19-affected patients develop mild to severe respiratory failure and heal without specific intervention. Older persons and individuals with serious medical problems such as cardiovascular disease, asthma, chronic respiratory disorders and cancer continue to experience extreme disease. Throughout the battle against coronavirus, there were several studies about the use of AI. A global overview of the deep learning strategies, which have been used until now, and the potential course of study, is quite relevant in the present scenarios. Thus, it is very much essential to study and analyse the AI techniques available in the literature to be utilized to assess COVID-19 patients. In this work, we have used deep learning strategies on both CT scan and X-ray images to assess COVID-19 patients.

The coronavirus disease or COVID-19 is a fast-spreading pandemic caused due to the SARS-CoV-2 virus causing the death of many peoples worldwide. The conventional methods of disease detection and diagnosis like swab test using RT-PCR are not sufficient enough during this critical condition as it has several limitations along with possibilities of being contaminated. Computer-based tools are now being used for the demonstration of the disease and healthcare management. The present chapter is to demonstrate the various applications of AI-based model that is useful against COVID-19, based on recently developed technologies and research publications. The AI-based algorithm is driven by machine learning technology along with an advanced bio-computational technique for fast and precise diagnosis and detection of coronavirus disease. It also has the ability of early prediction and warning for the spread of disease. Moreover, AI-based techniques are also an important setup for the development of an effective drug or vaccine. It provides worldwide access to various databases of all research and medical data related to COVID-19 and also helps in the management of the socioeconomic constraints. This study summarizes the application of the artificial intelligence-based model and its utilities in the fight against this pandemic, along with its limitations and future advancement and developmental strategies.

The number of confirmed cases of COVID-19 is increasing exponentially day by day across the world because of its super spreading nature. It was started in China and took a very less time to spread all over the globe. Due to its mortality rate, spreading nature, and unavailability of proper medicine and vaccination, it is declared as a pandemic by the World Health Organization (WHO) in March 2020. In this crisis time of the COVID-19 outbreak, technologists try to smooth the lives by minimizing the infection rate and facilitating in-time quality treatment. In this work, we collected the world data of COVID-19 cases in terms of confirmed, recovery, active, and death and provided visualization. We have also tried to find the patient’s risk level in terms of high, medium, and low by analyzing the patient’s symptoms and previous health histories such as high blood pressure, cardiac disease, diabetes, kidney issues, and others. We applied the C4.5 machine learning (ML) classifier to the considered dataset after preprocessing for risk assessment. The results obtained from the study indicate that the algorithm helps in achieving 75% accuracy.

An outbreak of coronavirus pneumonia was firstly documented in Wuhan, Hubei Province, China (December 2019), with an indication of human-to-human transmission. The causative agent identified for coronavirus disease 2019 (COVID-19) is the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). America, Italy, India, and Africa became new venues of COVID infection; the overall data of patients and death is increasing day by day. Generally inplace of most of the infected people develop respiratory symptoms (throat pain, cough, etc.), fever, and chest opacity on CT scan and X-ray. A few numbers of suspected persons are found asymptomatic; they may serve as carriers for infection. As a point of care, the patient diagnosis is compulsory, and only the diagnosis can provide a real-time condition of patients and can be helpful in arresting the spreading of the infection. In the present chapter, we focused on illustrating various diagnostic techniques that have been employed by the world for the detection of the coronavirus. The diagnostic techniques are categorized into molecular and serologic assay techniques. The nucleic acid is detected in molecular assay, whereas the serologic assay uses antigen-antibody reaction.

The vulnerable coronavirus (COVID-19) changed our lifestyle drastically and is considered as the highest threat to humanity. However, it proves there is an immense possibility for improvement in lifestyle of people. Social control is one of the main aspects to cease the spread of COVID-19. Manifold researches have been published in recent times for effective monitoring and management of social distancing in public/private places. Motivated by this notion, integrating embedded hardware kit like Jetson Nano with deep learning algorithms for automating the entire process is indispensable. Further, to improve the performance of deep learning-based object identifiers, YOLO (You Only Look Once) object detection algorithm utilizes a single-step locator methodology by computing pairwise distance to detect the objects. Using bounding boxes, YOLO algorithm identifies the presence of multiple objects, and thereby multiple bounding boxes are formed and update the shade of the jumping box to red. An alarm message is sent to the respective authority via WhatsApp. A real-time case study is detailed in this chapter for social control, thereby preventing the spread of COVID-19 with the aid of YOLO algorithm. In addition, specifications of Jetson Nano board and the environment setup are elaborated. It is observed from the experimental framework that effective monitoring and maintaining social control is possible to break the chain of spread of this contagious disease through YOLO object detector.

The most highlighted international slogan of the year 2020 is “Stay Home; Stay Safe.” The Internet of Things (IoT) has bridged the gap between the physical world and digital world in the COVID-19 outbreak. It provides a promising solution to the modern health issues with its technological, social, and economic prospects. Due to the rapid increase in confirmed cases through laboratory tests, this technological advancement helps the people to protect themselves from the infections. Even when the persons become the victims of COVID-19, the Internet of Things-based healthcare devices help in remote monitoring during quarantine, in order to prevent more infections among the public. However, numerous fake news are busting in a day-to-day basis about COVID-19, and security concerns on patient’s data are at risk. Artificial intelligence and blockchain technologies are explored to address these challenges further; a major review is outlined on various aspects such as Internet of Health Things, Internet of Medical Things, telemedicine, and Internet of Medical Things Industry Status. This chapter proposes a new category, “Internet of Covid Things,” along with their challenges. Mobile applications which will assist almost all people and the industry aspects on manufacturing the medical devices are also included in the review framework.

Many application problems are formulated as nonlinear binary programming models which are hard to be solved using exact algorithms especially in large dimensions. One of these practical applications is to optimally distribute protective materials for the newly emerged COVID-19. It is defined for a decision-maker who wants to choose a subset of candidate hospitals comprising the maximization of the distributed quantities of protective materials to a set of chosen hospitals within a specific time shift. A nonlinear binary mathematical programming model for the problem is introduced with a real application case study; the case study is solved using a novel discrete binary gaining-sharing knowledge-based optimization algorithm (DBGSK). The solution algorithm proposes a novel binary adaptation of a recently developed gaining-sharing knowledge-based optimization algorithm (GSK) to solve binary optimization problems. GSK algorithm is based on the concept of how humans acquire and share knowledge through their life span. Discrete binary version of GSK named novel binary gaining-sharing knowledge-based optimization algorithm (DBGSK) depends mainly on two binary stages: binary junior gaining-sharing stage and binary senior gaining-sharing stage with knowledge factor 1. These two stages enable DBGSK for exploring and exploitation of the search space efficiently and effectively to solve problems in binary space.

COVID-19 is an acute infectious respiratory novel coronavirus disease. The infection is caused by SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) and confirmed as a global pandemic by the World Health Organization. To combat the pandemic, it is desired to develop an effective and safe vaccine against SARS-CoV-2. In silico methods for vaccine development could be the alternative methods to conventional methods which are complex and time-consuming. In silico methods are effective, safe, and less time-consuming. This in silico method includes reverse vaccinology, immunoinformatics, and structural vaccinology. Recently, several vaccines have been developed by using bioinformatics tools, and they are under different phases of clinical trials. We have enlisted different tools and software for the in silico designing and validation of vaccine against SARS-CoV-2.

The COVID-19 pandemic has been a global health crisis since December 2019, when the first infection was reported in Wuhan, China. The critical and lethal advancement of this disease is associated with the failure of multiple organs including, but not limited to, the brain, lungs, heart, liver, kidneys, etc., which makes it very challenging to understand. Current high-throughput technologies generate multi-omics datasets to enable a comprehensive and in-depth analysis of different organs at the cellular and molecular level. To understand the multi-organ impact of COVID-19 and the mechanistic aspects of disease prognosis and its interactions with other comorbidities, computational approaches need to be implemented by integrating data from multiple organs, correlating results across data types, and applying machine learning (ML) tools on the high-throughput data.

This chapter is expected to provide valuable insights to help explain the multi-organ association of COVID-19 using state-of-the-art computational resources and modeling of high-volume data. We have emphasized the importance of big data analytics and systematic integration of data from different domains including omics, clinical, demographic, and others in understanding the organ- and system-level biological processes and comorbidity networks associated with COVID-19. These findings and proposed strategies could help perform comorbidity-focused studies to understand and tackle COVID-19.

Due to the outbreak of COVID-19 pandemic, there exists a situation of worldwide/nationwide lockdown across different countries which results in the growth of new-age technologies including augmented reality (AR) and virtual reality (VR) and their demands. Lockdown situation and social distancing had given sufficient time to develop innovative ideas by using AR/VR technologies. Many global seminars or conferences had been cancelled to practice social distancing and adopt the AR/VR technologies. The consequences of COVID-19 pandemic such as uncertain prognoses and shortages of resources had led to an increase in the depression level among people. This chapter demonstrates the applications of AR/VR and its proposed methodology to integrate the benefits of VR to the standardized mental health treatment process of mental health issues such as depression. The need for effective diagnosis and treatment of mental health combined with the recent advancement of artificial intelligence technologies had led to an increase in explorations in this direction. Deep learning has potentially discovered new learning patterns on the human-machine interface to identify risk factors of mental illness as well as to give optimized therapies.

In 2019, coronavirus infection hit Wuhan, China, and spread throughout South Korea, Italy, Iran, the USA, and the rest of the world. Medical scientists worked tirelessly to develop vaccines. WHO introduced preventive/protective measures to reduce human-human transmission, while researchers are searching for alternative technologies in AI, ML, and feature engineering to improve medical test accuracy, perfect isolation, and disease control. Recent research unraveled radiology imaging techniques that predict fast spread and have accurate diagnosis by confirming pathogens in blood cells. Machine learning techniques utilized to identify and diagnose COVID-19 are medic-tech applications for analyzing pneumonic effects of the virus in the body. ML approaches can help to examine the heart through chest X-rays to reveal how it affects the lungs, kidney, liver, and other vital organs in the body. X-ray accuracy can be analyzed with multinomial MNB, ANN, SVM, GANS, and other deep learning tools that validate scan results. Data authenticity above 85% may depend on computed algorithms that show whether COVID-19 nucleic acid test results are negative or positive. Polymerase chain reaction (PCR) provides a series of DNA samples for performing genetic test analysis. PCR helps experts to predict COVID-19 infectious agents using human DNA samples. However, PCR that supposes to show how COVID-19 spreads can be influenced by medical protocols causing further contaminations. Alternative use of technology has brought about an automated response while using CNN (convolutional neural network) to predict viruses and offer diagnostic options. Chest X-ray images can easily be classified using ML fractional multi-channel exponent moments (FrMEMS) to indicate infected patients. The relevant approach needed to achieve fractional multi-channels exponent process involves the adoption of parallel computational methods that accelerates, modifies, and optimizes Manta Ray Foraging to improve dataset accuracy from 85% to 98%. This report on machine learning techniques for identification and diagnosis of COVID-19, discussed ML performance techniques, identifying image process applications, image classification & analysis with ML, data segmentation, machine learning methods; that classifies x-ray scans, determine rates of COVID-19 spread, normal & abnormal cases, and recommend preventive measures.

The novel coronavirus SARS CoV-2 (COVID-19) is a pandemic disease and became a public health emergency worldwide. In the present study, comparison among most vigorously affected select cities across different countries, Delhi (India), Madrid (Spain), Lombardy (Italy), and New York and New Jersey (USA), were carried out up to July 2020. Predictive modelling was employed by using machine learning algorithms to predict the spread of the virus during August to November 2020. The results indicated that population density and urban density have a stronger connection with the spread of COVID-19 across the cities of Madrid, New York, and New Jersey. Conversely, Milan has exhibited a higher infection rate despite low population density (420 persons/km²) because of delayed human-human transmission measures and lockdowns. Relatively lower infection was recorded in Delhi even with higher population density (11,312 persons/km²) and higher urban compactness, which can be attributed to timely lockdowns and social distancing measures. The temperature and humidity have also abetted the spread of virus, especially in temperate regions with threshold temperature <10 °C with a humidity level between 60 and 77 g/m³. Predictive modelling reveals withdrawal of the pandemic across select cities by the end of 2020. The above findings would assist policymakers in making appropriate decisions for preventing the spread of this novel virus.

The coronavirus outbreak occurred in our current interconnected intelligence-crammed world. However, circulation of precise up to the minute information about the escalation of the disease is a persisting challenge. If there’s a faint hope to any of the chaos happening across the globe right now, it’s that people and technology are entwining in ways no one could have prognosticated. Chatbots, the conversational artificial intelligence (AI), are entering the struggle to keep the thousands of panicked individuals informed around the clock. A Chatbot is a software program, which is efficient in managing conversation by text or voice commands with people in natural language. Back in the day, Chatbots were no more than a fancy toy, although the spread of the coronavirus pandemic is contributing more value to Chatbots when it comes to taking on the tasks of overloaded hospitals by answering common but important medical questions asked by the consumers while needing to stay at home. The aim of this review is to explore the current availability and developments in Chatbots along with the necessity for virtual assessment tool or Chatbots in the field of healthcare and their role in detecting COVID-19 symptoms, query handling and myth bursting about the global pandemic threat, the coronavirus.

The novel pneumonia virus known as Covid-19 exceeds prior intervention, triggering the use of medical technology approach to counter the rise of coronavirus cases that spread uncontrollably around the globe. More than 2.5 million records of infected humans, animals, and sea lives are partially at the risk of causality. Biomedical research presently conducted with competent technological solutions such as AI, Cloud computing, big data, telemedicine, blockchain, and IoT were evaluated to review its authenticity in subsiding the effect of Covid-19. IoT wearable network and wireless microcontroller are used for contact tracing, biosensory point of care testing, and deploying of Internet of Medical Things to reduce spread of viral diseases. This report will focus on IoT relevance in alleviating Covid-19 issues, unique IoT applications crucial for fighting the pandemic, and its applicability. The objective of this research is to evaluate how IoT applications are implemented as smart tool for collecting Covid-19 symptoms, monitoring the spread of the virus, handling contact tracing, computing results using edge analytics, etc. It will also review Internet of Medical Things designed to combat different waves of infectious diseases, how to detect Covid-19 using IOT DSN and D2D, and the validation of IoT wearables through simulations.

Human civilization witnessed an array of disastrous events since the very beginning. The ongoing global pandemic is the latest addition to that. Indeed, the effects of the present one have been overwhelming across the vast global periphery. Since the onset of the Covid-19 pandemic, there has been a constant debate going regarding the potential origin. The most crucial part of the ongoing event without a doubt, however, has been the hallmark along with the pandemic itself is the negative “infodemic,” and that has been the most evident obstacle to counter, thus far. The fabricated version of the infodemic made the incidence several folds greater than it would have been. Electronic social media alongside few section of netizens have been the major contributors toward the eventual outcome. Therefore, it is of highest importance to pull the shocks up and wipe out the paranoid ideas to put the global healthcare system on the right track again.

The severe damage caused by COVID-19 has become a reality, and there is no longer a way to save humanity from this epidemic except diagnose and prevention, especially with emergence delay and lack of vaccine recognized by the World Health Organization. Without therapeutic treatment or explicit restorative immunizations for COVID-19, it is fundamental to diagnose the disease at an early stage and quickly seclude patients contaminated with the virus. This study aims at estimating the damage via consistency of chest imaging, which is not always feasible or possible. Here, an application is proposed to solve the problem via a WEB Predictor ‘COVIDz” and a program exploiting deep learning, so as emergency care will be able to systematically bring chest X-ray images and predict the percentage of the absence or presence of COVID-19. The proposed approach (custom VGG model) and our WEB site “COVIDz” objective validation of the suggested solution obtained the best classification efficiency of 99.64%, F-score of 99.2%, precision of 99.28%, MCC of 99.28%, recall of 99.28%, and a specificity value of 100%.