Application of Machine Learning and Deep Learning Techniques for COVID-19 Screening Using Radiological Imaging: A Comprehensive Review.

Lung, being one of the most important organs in human body, is often affected by various SARS diseases, among which COVID-19 has been found to be the most fatal disease in recent times. In fact, SARS-COVID 19 led to pandemic that spreads fast among the community causing respiratory problems. Under such situation, radiological imaging-based screening [mostly chest X-ray and computer tomography (CT) modalities] has been performed for rapid screening of the disease as it is a non-invasive approach. Due to scarcity of physician/chest specialist/expert doctors, technology-enabled disease screening techniques have been developed by several researchers with the help of artificial intelligence and machine learning (AI/ML). It can be remarkably observed that the researchers have introduced several AI/ML/DL (deep learning) algorithms for computer-assisted detection of COVID-19 using chest X-ray and CT images. In this paper, a comprehensive review has been conducted to summarize the works related to applications of AI/ML/DL for diagnostic prediction of COVID-19, mainly using X-ray and CT images. Following the PRISMA guidelines, total 265 articles have been selected out of 1715 published articles till the third quarter of 2021. Furthermore, this review summarizes and compares varieties of ML/DL techniques, various datasets, and their results using X-ray and CT imaging. A detailed discussion has been made on the novelty of the published works, along with advantages and limitations.

COVID-19 Radiography Using ConvNets

The COVID-19 pandemic continues to have a negative impact on the fitness and well being of the worldwide population. A vital step in tackling the COVID-19 is a successful screening of patients, with one of the key screening approaches being radiological imaging using chest radiography. This study aims to automatically identify patients with COVID-19 pneumonia using digital x-ray images of the chest while increasing the accuracy of the diagnosis using Convolution Neural networks (CNN). The data-set consists of 5380 X-ray images consisting of 1345 X-ray images each of COVID patients, Lung Opacity, Normal patients and Viral Pneumonia. In this study, CNN based model have been proposed for the detection of coronavirus pneumonia infected patients using chest X-ray radiography and gives a classification accuracy of 93.77% (training accuracy of 99.81% and validation accuracy of 95.45%). © 2022 IEEE.

COVID-19 prediction from chest X-ray images using deep convolutional neural network

One of the biggest challenges that the world is facing right now is the identification of COVID-19 infection, given no potential vaccine for the fast-spreading virus. Ongoing insights demonstrate that the number of individuals infected with COVID-19 is expanding exponentially, with more than 40 million confirmed cases around the world. One of the pivotal steps in battling COVID-19 is the capacity to recognize the infected patients sufficiently early and put them under isolation. One of the quickest approaches is to predict the illness from radiography and radiology pictures. Propelled by prior works, I present a machine learning binary classification model-driven deep convolutional neural network to predict COVID-19 from chest X-Ray images. A blend of Dr. Joseph Paul Cohen’s open-sourced database and Kaggle’s Chest X-ray competitions dataset were used to train our model. The predictions result of the model exhibit promising performance with an accuracy of 95.61%. Training and validation accuracy graphs along with training and validation loss graphs were plotted for a better comprehension of our model. Further evaluation of the model was done by calculating standard evaluation metrics where 100% sensitivity, 93.33% specificity, 93.75% precision, and F1-score of 96.77% were achieved. The results exhibit that advanced machine learning methods combined with radiological imaging proved to be a deployable methodology for correct diagnosis of COVID-19, and can likewise be assistive to defeat the issues like shortage of testing kits, time-consuming, and expensive testing methods. © 2022 Elsevier Inc. All rights reserved.

Detection of COVID-19 using CNN from Chest X-ray Images

COVID-19 (novel coronavirus disease) is a deadly illness, has infected and killed a very large number of people worldwide. The widely followed lab testing (RT-PCR Test) for the detection of this disease has various limitations with high cost and take long time to provide the outcome. As a result, diverse technologies that permit for the quick and accurate finding of the infection can provide much required assistance to medical management. In recent studies, gained radiological imaging techniques, such images convey important information about this virus. Advanced Deep learning (DL) techniques combined with the radiology images can aid in the correct diagnosis of the virus, as well as defeat the problem of insufficient expert physicians in rural areas. In this work, aimed at presenting a DL based-Convolutional neural network (CNN) model for the automatic detection of the coronavirus from X-ray images of chest. The Kaggle dataset available publicly of total 42330 images from 4-categories are used. The experiment produced the accuracy of 88.53% and 86.19% for training and validation, which is better result for the highest number of radiographic images in comparison to existing work. © 2022 IEEE.

A Survey of Deep Learning on COVID-19 Identification Through X-Ray Images

New Coronavirus 2019 (COVID-19) is a virus that causes severe pneumonia and affects many organs of the body. This infection was initially discovered in one of the cities in the Republic of China, Wuhan, in December 2019 and since then has been spread throughout the globe as a global pandemic. To prevent the virus from spreading, positive cases must be identified early and infected persons must be treated as soon as possible. As new instances emerge regularly, many developing countries are experiencing COVID-19 testing kit scarcity because the demand for testing kits has soared. As an alternative, radiological imaging techniques such as X-ray images have been proven to help in COVID-19 diagnosis because images from X-ray provide valuable information about the COVID-19 virus disease. This paper presents a survey of Deep learning-based methods in identifying COVID-19 with X-ray input images, and classifies these images into several categories, namely: no findings, normal, COVID, and pneumonia. Several studies have been included with details about their datasets, methodologies, and findings. A total of thirteen popular datasets and fifteen articles are reviewed in this paper. Research challenges and recommendations for future research directions are also provided as an evaluation of previous research. Search for research articles in well-known digital libraries, namely Scopus, IEEE Xplore, Springer, and ScienceDirect, was carried out to obtain a list of studies relevant to the scope of research. Related articles that have a high impact are considered in the list of studies. Also, in selecting studies related to the research scope, we apply some inclusion and exclusion criteria. The list of studies used in subsequent research is imported to the library. Then, studies that did not match the criteria for inclusion were eliminated. The clinical application of artificial intelligence, i.e., DL in diagnosing COVID-19, is promising, and further research is needed. Convolutional Neural Network (CNN) approaches could be used in collaboration through X-ray pictures to identify diseases quickly and accurately, reducing the shortage of testing equipment and their restrictions. It is expected that this work can help researchers understand the general picture and existing research gaps to decide on the appropriate architecture and approach in developing deep learning-based covid identification research. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Automated detection of COVID-19 from chest X-rays using CNN

The novel coronavirus (COVID-2019), which first arrived in the Chinese city of Wuhan in December 2019, unrolled like a wind around the arena and brought into an epidemic and declared as a worldwide pandemic by WHO in march,2020. Agriculture, building, manufacturing, trade, lifestyle, tourism, and the global economy all suffer as a result of this disease. Consequently, it's very crucial to diagnose and treat the disease as soon as possible. According to radiology imaging methodologies radiological imaging techniques may aid in appropriately diagnosing and treating the condition with less response time. The utilization of raw chest X-ray pictures has been used to introduce a new model for the automatic detection of COVID-19 in this investigation. In this work we have built a binary classifier to detect covid-19 by deploying the deep learning technique-CNN on dataset collected from repository of JHU CSSE 2019 and was supported by JHU APL and transfer learning techniques has also been utilized to enhance the dataset. The best classification accuracy achieved by our model on chest X-ray dataset is 98.3%. We have also analyzed and compared the model in other research papers with our model. © 2022 IEEE.

Detection of COVID-19 Cases from Chest Radiography Images

The COVID-19 epidemic continues to have a devastating influence on the global population's well-being and economy. One of the most important advances in the fight against COVID-19 is thorough screening of infected individuals, with radiological imaging using chest radiography being one of the most important screening methods. Early studies revealed that patients with abnormalities in chest radiography images were infected with COVID-19. Persuaded by this, a variety of computerized reasoning and simulated intelligence frameworks based on profound learning have been suggested, with promising results in terms of precision in differentiating COVID-infected individuals. COVID-Net, a neural system configuration custom-fit for the recognition of COVID-19 instances from chest radiography photographs that is open source and accessible to the general public, is presented in this study. Many techniques have been used for the detection of COVID-19, but here we are going to focus on the chest radiography technique with the application of machine learning and image processing concepts. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Post COVID-19 fitness to dive assessment findings in occupational and recreational divers.

INTRODUCTION: It is now known that COVID-19 has long term effects that may not correlate with clinical severity of disease. The known pulmonary and cardiovascular changes as well as thrombotic tendency could predispose to diving accidents. We aimed to investigate COVID-19 related changes that may cause disqualification from diving among divers who recovered from the disease. METHODS: Occupational and recreational divers who applied for fitness to dive (FTD) assessment after COVID-19 infection were included. Routine FTD assessments were performed. Details of COVID-19 history were evaluated. Lung computed tomography (CT) scans were advised if not previously performed or if there were COVID-19 related changes in previous scans. Divers with pathological findings were restrained from diving and followed prospectively. RESULTS: Forty-three divers were analysed. Thirteen divers were restrained from diving, all due to persistent COVID-19 related changes in lung CT. The prevalence of CT with at least one lung lesion was 68.2% at the time of diagnosis, 73.3% in the first three months after diagnosis and 19.2% later. The most common CT findings were glass ground opacities and fibrotic changes. Demographic characteristics and COVID-19 history of divers deemed 'unfit' were similar to those deemed 'fit'. CONCLUSIONS: Divers who recover from COVID-19 should undergo FTD assessments before resuming diving. A chest CT performed at least three months after diagnosis may be suggested.

COVID-19 and pulmonary fibrosis: therapeutics in clinical trials, repurposing, and potential development.

In 2019, an unprecedented disease named coronavirus disease 2019 (COVID-19) emerged and spread across the globe. Although the rapid transmission of COVID-19 has resulted in thousands of deaths and severe lung damage, conclusive treatment is not available. However, three COVID-19 vaccines have been authorized, and two more will be approved soon, according to a World Health Organization report on December 12, 2020. Many COVID-19 patients show symptoms of acute lung injury that eventually leads to pulmonary fibrosis. Our aim in this article is to present the relationship between pulmonary fibrosis and COVID-19, with a focus on angiotensin converting enzyme-2. We also evaluate the radiological imaging methods computed tomography (CT) and chest X-ray (CXR) for visualization of patient lung condition. Moreover, we review possible therapeutics for COVID-19 using four categories: treatments related and unrelated to lung disease and treatments that have and have not entered clinical trials. Although many treatments have started clinical trials, they have some drawbacks, such as short-term and small-group testing, that need to be addressed as soon as possible.

Deep Learning Approaches for Detecting COVID-19 From Chest X-Ray Images: A Survey.

Chest X-ray (CXR) imaging is a standard and crucial examination method used for suspected cases of coronavirus disease (COVID-19). In profoundly affected or limited resource areas, CXR imaging is preferable owing to its availability, low cost, and rapid results. However, given the rapidly spreading nature of COVID-19, such tests could limit the efficiency of pandemic control and prevention. In response to this issue, artificial intelligence methods such as deep learning are promising options for automatic diagnosis because they have achieved state-of-the-art performance in the analysis of visual information and a wide range of medical images. This paper reviews and critically assesses the preprint and published reports between March and May 2020 for the diagnosis of COVID-19 via CXR images using convolutional neural networks and other deep learning architectures. Despite the encouraging results, there is an urgent need for public, comprehensive, and diverse datasets. Further investigations in terms of explainable and justifiable decisions are also required for more robust, transparent, and accurate predictions.

Artificial Intelligence Applied to Chest X-Ray Images for the Automatic Detection of COVID-19. A Thoughtful Evaluation Approach.

Current standard protocols used in the clinic for diagnosing COVID-19 include molecular or antigen tests, generally complemented by a plain chest X-Ray. The combined analysis aims to reduce the significant number of false negatives of these tests and provide complementary evidence about the presence and severity of the disease. However, the procedure is not free of errors, and the interpretation of the chest X-Ray is only restricted to radiologists due to its complexity. With the long term goal to provide new evidence for the diagnosis, this paper presents an evaluation of different methods based on a deep neural network. These are the first steps to develop an automatic COVID-19 diagnosis tool using chest X-Ray images to differentiate between controls, pneumonia, or COVID-19 groups. The paper describes the process followed to train a Convolutional Neural Network with a dataset of more than 79, 500 X-Ray images compiled from different sources, including more than 8, 500 COVID-19 examples. Three different experiments following three preprocessing schemes are carried out to evaluate and compare the developed models. The aim is to evaluate how preprocessing the data affects the results and improves its explainability. Likewise, a critical analysis of different variability issues that might compromise the system and its effects is performed. With the employed methodology, a 91.5% classification accuracy is obtained, with an 87.4% average recall for the worst but most explainable experiment, which requires a previous automatic segmentation of the lung region.