System Design for Epidemics Using Machine Learning and Deep Learning

Mutation in viruses is known to be an unavoidable phenomenon. But at times, it may become a life-threatening pandemic just like in the case of the 2019 novel coronavirus, formally named as SARS-CoV-2, which consumed around 36,405 lives out of 750,890 infections as per the data available with the World Health Organization as of the end of March 2020. Found to be from the family of earlier known outbreaks (SARS and MERS) of the twenty-first century, it has now become a public health emergency of international concern (PHEIC). Countries around the world have spent millions of dollars to get a positive sign of finding vaccines, but still it remains an unsolved mystery. Even though there is implementation of strict lockdown measures from several affected countries around the globe, the trend line of COVID-19 epidemic is still increasing exponentially. Being in this scenario, this paper deals about the outbreak of 2019-nCoV and its structure, growing stages, global statistics, transmission modes, and most possible precautionary methods and also its emphasis on creating public awareness by answering few key clarifications about novel beta coronavirus disease. The machine learning method used in this study was taught using records from COVID-positive tests. Results from a week were included in the testing set (individuals who were confirmed to have COVID-19). This proposed model predicted the COVID-19 lab findings with high accuracy by only using eight numeric data, age 60, knowing contact with an infected individual, and the existence of five early clinical signs.

Managing the updated medical history and records of healthcare is a comprehensive process. With technology like Vision API, AutoML Google Cloud has done a remarkable role in data storage and retrieval. As we all know that the manual data process, paperwork, and similar kind of documentation have extensive processes that usually take huge time, effort to complete the majority of the processes takes a lot of time and money to complete. So, to reduce the time and efforts at a low cost, now we have an excellent technique to cope up with all these issues named machine learning algorithms. Vector machines and ML-based OCR recognition techniques are deliberately collecting potency in objects and different document classification methods such as Google’s Cloud Vision API and MATLAB’s machine learning-based handwriting recognition technology. Researchers of MIT are innovating new techniques with the help of machine learning which is the succeeding generation of astute, smart health records, by integrating ML-based tools from scratch to encourage better analysis and diagnosis, and superb treatment recommendations.

The coronavirus disease (COVID-19) makes humans suffer from mild to moderate respiratory problems, with severe cases requiring special treatment. In many severe cases, elderly individuals and people with pre-existing medical issues like lung-related disease, insulin-dependent disease, and carcinoma, are more prone to difficulty breathing and developing a severe illness. To detect the coronavirus here, X-ray radiograph images are considered. The main motive for using X-ray radiograph images is their being cost-effective and being able to give considerable accuracy compared to its counterpart, computed tomography (CT) scans. In this study, the deep learning model Visual Geometry Group (VGG)16 using the transfer learning method and image augmentation techniques was employed for automatic COVID-19 diagnosis. These two techniques will assist the deep learning model to learn the target task by improving the baseline performance by using fewer X-ray radiograph images in the training phase and showing improvements in the model development time by utilising knowledge gained from a source model. Many deep learning methods have been published in the literature to solve the same cases, but the proposed method uses a simple VGG16 model with transfer learning, which takes less processing time and gives satisfactory results even by using fewer training samples.

The use of AI and ML is considered as a boon for several industries such as healthcare, manufacturing, robotics and gaming to combat rising competition and improve service quality, efficiency and reaction time. The United Nations has been encouraging every industry to contribute to the Sustainable Development Goals (SDGs) since 2015. These goals were set to be achieved by 2030, taking into account all aspects of life. The progress of AI over the last decade has greatly aided these SDGs in numerous ways. Synchronised and integrated AI use in multiple areas can help governments achieve these SDGs more efficiently. Some SDGs, like good health, quality education, clean energy and economic growth, are directly related to AI, while others are indirectly related to AI. In India, the circular economy and smart cities are top priorities. As of 2020, 79% of SDGs have been accomplished, with AI helping to create ‘smart cities’ and ‘circular economies’. The everyday routine difficulties such as managing medical resources, traffic congestion, weather prediction, free parking slots and taxi and cab arrival time are well controlled by AI, contributing to time management and environmental contribution. This way we can see AI’s role in SDG achievement. This chapter intends to highlight the role of AI in several SDGs for the national economy by highlighting their direct contribution in various sectors and social systems.

IoT technology is emerging as a fully developed automation that could be integrated in various web applications, which will be present in upcoming generations of the World Wide Web. Blockchain, like IoT, is a burgeoning field whereby every system associated in the blockchain incorporates a disseminated ledger that improves safety and consistency. Due to the blockchain network abilities to accomplish smart contracts and consensus, unauthorized users are unable to undertake any fault transactions. The IoT and blockchain can be aggregated to improve application performance dynamically at run time. However, controlling and monitoring the machines linked to sensors in an IoT background and mining the blockchain will always be a technical challenge to the researchers. With this context, this paper enables to review the fundamentals of IoT, blockchain field, and its topographies. In this paper, design architecture, namely, IoT Blockchain Assurance-Based Compliance to COVID Quarantine, is proposed and concluded up with novel architectural framework that improves the efficiency of data safety and data transparency. Unlicensed users are not permitted to conduct any erroneous transactions within the blockchain network, which has the capability to engage in smooth contracts and agreement, thus extending the safekeeping between clinicians and chronically ill patients. This methodology was created with immobile elderly chronically ill patients in mind who are suffering from COVID that require on-the-spot treatment and continuous monitoring by a doctor in mind. This paper is designed to analyze the performance of proposed IoT Blockchain Assurance-Based Compliance to COVID Quarantine with Ethereum private blockchain network beneath a genesis block and the results are conferred.

A brain tumour is caused by an unregulated rise as concerns brain tissue with aberrant cells. There are two types of brain tumours: malignant and benign tumours. Early discovery of a brain tumour may be necessary to ensure a patient’s survival. MRI scanning is typically used to detect brain tumours. However, because of the uneven shape of tumours and their location in the brain, radiologists are unable to provide good tumour segmentation in MRI images. The method of using MRI to segment and classify an infected tumour region by processes of segmentation, detection as well as extraction is a big issue and time-consuming task that can only be completed by medical doctors with extensive knowledge. As a result, it is essential to overcome this by computer-aided technology (CAD). Brain MRI scans are particularly useful for determining difficulties such as persistent weakness, frequent headaches, hazy vision and faintness, as well as identifying definite persistent nerve disease such as multiple scleroses. To answer the problem of brain tumour segmentation from MRI images, this work proposed a deep learning technique for brain tumour segmentation that combined CNN with random forest. This method includes operations such as pre-processing, feature extraction, picture categorisation and brain tumour segmentation. As a consequence, the photographs are classified using random forest depending on the attributes chosen. A kernel-based CNN technique is used to segment the tumour from an MRI picture. When compared to existing algorithms, the proposed method’s experimental results show that it can accurately segment brain tumours with an accuracy of 95.4 per cent.

COVID-19 is an infectious disease caused by SARS-CoV-2 virus. It has disrupted the normal life of people, medical infrastructure, and economy globally. Remote health monitoring is a better option in pandemic diseases such as COVID and Ebola virus. Remote health monitoring can be enhanced by effectively using various recent advancements in technology. Technological advancements such as Wireless Body Area Networks (WBAN), Internet of Things (IoT), Artificial Intelligence (AI), and medical robotics for improving the effectiveness of remote health monitoring in COVID-19 pandemic are reviewed and presented in this chapter. Building expert systems using WBAN, IoT, AI, and robotics is an optimal choice to remote monitor COVID and reduce infection spread and mortality. Detailed architecture, use cases, impacts, workflow, applications, and future directions toward building a better expert system is highlighted in this chapter.

The large-scale outbreaks of infectious pandemic diseases emerged regularly throughout history and created notable economic, social, and political disruptions. Major pandemics affect a wide geographic area significantly increasing morbidity and mortality. The world has come across numerous remarkable pandemics such as the Black Death, measles, smallpox, influenza, plague, cholera, Spanish flu, severe acute respiratory syndrome coronavirus (SARS-CoV), Middle East respiratory syndrome coronavirus (MERS-CoV), human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS) and Ebola virus and is now combating the new coronavirus disease 2019 (COVID-19) pandemic affecting humanity greatly. Studies suggest that the likelihood of pandemic threats is due to the diversity of pathogens, changes in the dynamics of disease transmission and severity, human-pathogen interaction, increased globalization, urbanization, huge exploitation of land and natural resources, and global warming.

The pandemic risk burden poses serious challenges to humanity and these trends will prolong and intensify over time. For the well-being of humanity, administration of public health measures, techniques to intercept and control infection, pharmaceutical intervention, global surveillance programs, novel technologies to identify disease biomarkers, and vaccine production prove to be effective beneficiary responses to identify and limit emerging outbreaks and to escalate preparedness and health capacity. The extensive amount of data produced during the pandemic has given a lot of chances to the researchers and healthcare providers to evaluate new trends, detect vulnerable groups, and solve long-standing issues in the healthcare industry. The healthcare industry has sought to use the most comprehensive data and predictive analytics software tools employing intelligent data technology, artificial intelligence (AI), machine learning (ML), and deep learning (DL) and has leveraged to gain insight, establish innovative ways to ease sustainable demand and supply, and pitch straight into the prospective benefits to foster the fight against the pandemic.

Hence, these predictive models can support hospitals, healthcare settings, state health organizations, and government establishments to speculate the influence of COVID-19 and prepare for the future. In this chapter, a comprehensive investigation of various data analytic tools that are used in expert systems, proposed for pandemic and epidemic diseases, is discussed. The key issues, challenges, and opportunities of the existing and current methods are also discussed.

In this chapter, we proposed a deep convolutional generative adversarial network (DCGAN) for producing augmented images on the PatchCamelyon (PCam) dataset to improve the metastatic tissue detection performance in the lymph node section using scanned images. Training of the DCGAN is performed on graphics processing unit (GPU) up to 300 training epochs. The DCGAN model increased the training images from 262,144 to 300,000 images. Frechet Inception Distance (FID) method is used to evaluate the fidelity and diversity of the DCGAN model. The augmented images which are generated from the most DCGAN are added to the training data of the metastatic tissue detection model. A VGG19 and InceptionV3 networks were used to build the classification models. The classification accuracy of the VGG19 and InceptionV3 models on the test data are 97.73 and 99.26 percentages, respectively. The classification performance of the model which is trained on the augmented dataset is much higher than the model trained on the original dataset. These extensive results show the significance of the DCGAN in data augmentation.

The scientific and technical sectors saw the necessity for innovative solutions as the effects of COVID-19 on society became clear. More basic research endeavors with long-term and significant effects have focused on the creation of novel diagnostics and the acceleration of vaccines. Researchers from many walks of life got together to tackle this problem in a truly global effort. In the medium term, efforts have focused on repurposing current technologies and utilizing additive manufacturing techniques to overcome shortages in safety equipment and disinfection. The development of innovative diagnostics and the acceleration of vaccines have been the focus of more basic research initiatives with an impact in the middle and lengthy. As a vital technology, photonics has supported all efforts, both directly and indirectly, to combat this type of pandemic. This viewpoint will provide an outline of the crucial part the photonics society played in the COVID-19 pandemic and talk about how the photonics society could assist in preventing future pandemic viruses.

COVID-19 has been a major issue among various countries, and it has already affected millions of people across the world and caused nearly 4 million deaths. Various precautionary measures should be taken to bring the cases under control, and the easiest way for diagnosing the diseases should also be identified. An accurate analysis of CT has to be done for the treatment of COVID-19 infection, and this process is complex and it needs much attention from the specialist. It is also proved that the covid infection can be identified with the breathing sounds of the patient. A new framework was proposed for diagnosing COVID-19 using CT images and breathing sounds. The entire network is designed to predict the class as normal, COVID-19, bacterial pneumonia, and viral pneumonia using the multiclass classification network MLP. The proposed framework has two modules: (i) respiratory sound analysis framework and (ii) CT image analysis framework. These modules exhibit the workflow for data gathering, data preprocessing, and the development of the deep learning model (deep CNN + MLP). In respiratory sound analysis framework, the gathered audio signals are converted to spectrogram video using FFT analyzer. Features like MFCCs, ZCR, log energies, and Kurtosis are needed to be extracted for identifying dry/wet coughs, variability present in the signal, prevalence of higher amplitudes, and for increasing the performance in audio classification. All these features are extracted with the deep CNN architecture with the series of convolution, pooling, and ReLU (rectified linear unit) layers. Finally, the classification is done with a multilayer perceptron (MLP) classifier. In parallel to this, the diagnosis of the disease is improved by analyzing the CT images.

Managing sensory data captured in leveraging is a challenge, especially during a pandemic when trying to capture the psychological, emotional, and physiology standards. The advanced technology of edge computing and IIoMT together help to reach promising outcome results from the home environment using psychological feelings and somatic health equivalent data. The basic application of Deep Learning leads to the asset-constraint of edge computing, which provides a way to move the data that is collected from IIoMT devices to various locations. All kinds of data related to health can exist in a particular place of user edge while assuring the security, privacy, and low latency of the inference system. In this article, an Internet of Medical system is developed that uses Deep Learning to detect risky types of health-related symptoms and generates reports and alerts for pandemic and epidemic situations, which helps in decision-making support. In these pandemic and epidemic situations, a lot of applications have been identified and implemented with their descriptions for the upcoming support for the real-time trials. We have developed smart applications in edge computing manuals. The overall output clearly allows us to view the fixed smart systems during the pandemic with the Smart Health Management system (SHMs).

Artificial intelligence (AI) has shown an immense potential to affect diverse domains of healthcare during the COVID-19 pandemic. The applications of AI in the field of cardiovascular disorders during the COVID-19 pandemic were an added advantage to the cardiologists, as it helped in certain aspects of treatments digitally. This technology is constructive for providing sophisticated treatment in the area of cardiovascular medicine based on technology, because it may assist in assessing and measuring the human heart function. Artificial intelligence employs simulated neuronal program for predicting the survival of a COVID-19 patient affected with heart dysfunction. AI entails intricate algorithms for predicting successful evaluation and therefore the treatment protocol. AI utilizes various methods like cognitive computing, deep learning, and machine learning. It is integrated to make an assessment and determine multifaceted challenges. In humans, cardiovascular disease is still one of the major causes of death, and it is escalating for years together and is also very expensive. AI is employed to recognize new drug treatment and advance the efficacy of a clinician. AI is turning into a well-approved attribute of a variety of engineering and healthcare segments and is being expected to provide a feasible treatment stage.

Precautionary measures are the best conceivable ways to impede the spreading of COVID-19 disease. Initial stage detection, proper analysis, suitable confinement, effective therapy and prompt vaccination are the key consideration to inhibit expedite transmission. Precautions are primarily concentrated on effective health screening, efficient treatment, and vaccination on time for each and every individual. Before the invention of COVID-19 vaccine, proper health checking or explorations are of predominant concern from therapeutic viewpoint as avoidance is superior than healing. At present scenario, impressive health screening through RT-PCR test, rapid antigen test, etc. is very imperative to identify Corona-positive cases in early stages even there is no disorders or symptoms of COVID-19 infection. Faster vaccination of most of the people, irrespective of cast, religion, and economic conditions, is indispensable to prohibit the transportation of such growing disorder and minimize its adverse effects. Recent literature reveals that most of the vaccines are safe and effective against coronavirus. A potent and competent vaccine lessens mild to moderate and serious conditions of COVID-19 patients regardless of comorbidities.

The population’s vulnerability is exacerbated by the lack of effective treatment drugs and immunity to COVID-19. The only viable strategy for combating this pandemic is social separation. In order to automate the task of monitoring social separation using surveillance footage, this study presents a neural network-based crowd density estimation for COVID-19 and future pandemics. The suggested framework employs the object identification model to distinguish persons in the scene, as well as the deep sort technique to track recognized people with issued IDs. The obtained results of the proposed work are compared in terms of loss values defined by object classification and localization, frames per second (FPS), and mean average precision (mAP). The proposed method yields good results against faster region-based convolutional neural network (RCNN) and single-shot detector (SSD).

The pandemic turned life upside down, including causing unavailability and an inability to access rehabilitation in the hospital. However, the need to be fit and healed does not stop, so rehabilitation innovation from the digital sectors plays a role in approaching the patient, as the patient requires a medical professional to be healed. Rehabilitation via a digital pathway is fraught with difficulties, but advances in technology and research have enabled it to be used to the greatest extent possible in this disaster. Digital health has increased its efficacy in response to the pandemic, as it is now available in developing countries where there is an inability to visit a clinic for rehabilitation, and now the rehabilitation tool is accessible to the patients in their hands and they can connect to their therapist at any time. The rehabilitation is designed based on the patient’s illness, feedback, and health data stored on the application devices, which regulate and provide feedback from both sides, from the patient and other improvement changes gathered with the help of digital applications. Digital health allows for online consultation, assessment, and 24-h monitoring, all of which are directly shared with the rehabilitation team.

As the population continues to age, the world is facing an epidemic of devastating neurodegenerative diseases that take a tremendous economic and emotional toll on families and society. Neurological disorders affect the brain and the nerves throughout the body, and they also affect the spinal cord. According to the World Health Organization (WHO), worldwide one million people are affected by different types of neurological disorders and count for seven million deaths. India has approximately 18% of the total world population; of this, around 3% of the population is affected by various neurological disorders. The most common neurological disorders are epilepsy, seizures, stroke, Parkinson’s disease, sclerosis, and Alzheimer’s disease. Various artificial intelligence algorithms have been devised to assess neurological illnesses as the field of neuroimaging data processing has grown rapidly. Computer-aided diagnosis (CAD) uses machine learning to deliver reliable information for the diagnosis of a patient’s ailment. Supervised, unsupervised, and reinforcement learning are the three forms of learning in machine learning (ML) approaches. After the training features of unlabeled datasets are found, supervised learning employs the labeled data to train an algorithm. Patterns and classes in the datasets without labels are learned by using unsupervised learning. This chapter will discuss how to diagnose and interpret key neurological conditions such as epilepsy, seizures, stroke, Parkinson’s disease, sclerosis, and Alzheimer’s disease using effective supervised and unsupervised learning techniques.

Due to sudden evolution and spread of COVID-19, the entire community in the globe is at risk. The covid has affected the health and economy and caused loss of life. In India, due to social economic factors, several thousands of people are infected, and India is seen as one of the top countries seriously impacted by the pandemic. Despite of having a modern medical instruments, drugs, and technical technology, it is very difficult to contain the spread of virus and save people from risk. Healthcare system and government personnel need to get an insight of covid outbreaks in the near future to decide on stepping up the healthcare facilities, to take necessary actions and to implement prevention policies to minimize the spread. In order to help the government, this study aims to build model a forecast COVID-19 model to foretell growth curve by predicting number of confirmed cases. Three variant models based on long short-term memory (LSTM) were built on the Indian COVID-19 dataset and are compared using the root mean squared error (RMSE) and mean absolute percentage error (MAPE). The findings have revealed that the proposed stacked LSTM model outperforms the other proposed LSTM variants and is suitable for forecasting COVID-19 progress in India.