Developing a transfer risk assessment tool for the Midlands Critical Care Network

Introduction: Inter-facility critical care transfers are a high-risk activity, with a significant reported critical incident rate.1 The 2019 ICS Transfer of the Critically Ill Adult Patient guideline2 recommends a consultant-led risk assessment is performed in order to provide a rationale for the make-up of the transfer team. Prior to our project, there was no formalised risk assessment process at our unit. Objective(s): We wished to assess whether any 'informal' risk assessment process was already being performed prior to transfers. We then aimed to implement a clear assessment process, initially for our unit but ultimately for our critical care network. Method(s): We performed a baseline audit of adult inter-facility critical care transfers undertaken by a team from our unit between 1st December 2019 and 28th February 2020. Notes were analysed for evidence of any risk assessment performed in discussion with the responsible consultant We then locally piloted a new risk assessment tool for our Critical Care Network's transfer documentation. It included the required elements from ICS guidance, and followed a systems-based approach to facilitate completion in time-critical situations. Colour coding enabled easy identification of potential high-risk transfers and guided team formation. Initial re-audit of the new tool was performed between 16th September and 16th October 2020, after which it was implemented across the network. A further re-audit was performed between 1st October and 31st December 2021. Result(s): Fifteen transfers occurred during the initial audit period. All were clinical. No risk assessments were documented (0% compliance), although all were accompanied by a transfer-trained, airway competent doctor and all but one by an ODP. Our second audit cycle identified 10 transfers, of which 4 had risk assessments completed (40% compliance). All transfers had been undertaken with a dual doctor/ODP team. We identified that there was limited knowledge of the risk assessment process among clinicians, so introduced the topic into our unit's transfer training programme. Assessment completion was made a key performance indicator, fed back to team members following each transfer. Our final cycle covered 14 clinical transfers. Eight had a fully completed risk assessment (57% compliance), 2 had partially completed risk assessments (14% partial compliance), 4 had no risk assessment and 2 cases were excluded due to incomplete data. Conclusion(s): Our tool is now used for all inter-hospital transfers across the Midlands Critical Care Network. It enabled risk assessments to be performed appropriately for transfers originating from our unit. Introduction was initially hampered by limited training for clinicians during the first wave of the Covid pandemic, and compliance improved once this was implemented. The recent introduction of a regional critical care transfer service means that the majority of transfers undertaken by our unit's staff are now time-critical clinical transfers. This may contribute to the failure to complete risk assessments in some cases, however these assessments are likely to be of higher importance since such transfers may be higher risk. We now aim to collect feedback from transferring staff to identify any barriers to correct completion.

Highly accurate multiclass classification of respiratory system diseases from chest radiography images using deep transfer learning technique

Chest X-ray radiographic (CXR) imaging aids in the early and accurate diagnosis of lung disease. The diagnosis process can be automated and accelerated by analyzing chest CXR images with artificial intelligence tools, particularly Convolutional Neural Network (CNN). Due to few medical images have been labeled, the most significant obstacle is utilizing these images accurately for diagnosing and tracking disease progression, and accordingly, the difficulty of automating the classification of these images into positive and negative cases. To address this issue, a deep CNN model was proposed to classify respiratory system diseases from X-ray images using a transfer learning technique based on the EfficientNetV2 model that acts as a backbone to enhance the efficacy and accuracy of Computer-Assisted Diagnosis (CAD) performance. Moreover, the latest data augmentation methods and fine-tuning for the last block in the convolutional base have also been carried out. In addition, Grad-CAM is used to highlight the important features and make the deep learning model more comprehensible. The proposed model is trained to work on the triple classification, COVID-19, normal, and pneumonia. It uses CXR images from three publicly accessible datasets. The following performance was achieved on the testing set: sensitivity = 98.66 %, specificity = 99.51 %, and accuracy = 99.4 %. Thereby, the proposal outperforms the four most recent classification techniques in the literature.Copyright © 2023 Elsevier Ltd

Rapid detection of COVID-19 from chest X-ray images using deep convolutional neural networks

The entire world is suffering from the corona pandemic (COVID-19) since December 2019. Deep convolutional neural networks (deep CNN) can be used to develop a rapid detection system of COVID-19. Among all the existing literature, ResNet50 is showing better performance, but with three main limitations, i.e.: 1) overfitting;2) computation cost;3) loss of feature information. To overcome these problems authors have proposed four different modifications on ResNet50, naming it as LightWeightResNet50. An image dataset containing chest X-ray images of coronavirus patients and normal persons is used for evaluation. Five-fold cross-validation is applied with transfer learning. Ten different performance measures (true positive, false negative, false positive, true negative, accuracy, recall, specificity, precision, F1-score and area under curve) are used for evaluation along with fold-wise performance measures comparison. The four proposed methods have an accuracy improvement of 4%, 13%, 14% and 7% respectively when compared with ResNet50.

An efficient lung disease classification from X-ray images using hybrid Mask-RCNN and BiDLSTM

Lung diseases mainly affect the inner lining of the lungs causing complications in breathing, airway obstruction, and exhalation. Identifying lung diseases such as COVID-19, pneumonia, fibrosis, and tuberculosis at the earlier stage is a great challenge due to the availability of insufficient laboratory kits and image modalities. The rapid progression of the lung disease can be easily identified via Chest X-rays and this serves as a major boon for the terminally ill patients admitted to Intensive Care Units (ICU). To enhance the decision-making capability of the clinicians, a novel lung disease prediction framework is proposed using a hybrid bidirectional Long-Short-Term-Memory (BiDLSTM)-Mask Region-Based Convolutional Neural Network (Mask-RCNN) model. The Crystal algorithm is used to optimize the scalability and convergence issues in the Mask-RCNN model by hyperparameter tuning. The long-range dependencies for lung disease prediction are done using the BiDLSTM architecture which is connected to the fully connected layer of the Mask RCNN model. The efficiency of the proposed methodology is evaluated using three publicly accessible lung disease datasets namely the COVID-19 radiography dataset, Tuberculosis (TB) Chest X-ray Database, and National Institute of Health Chest X-ray Dataset which consists of the images of infected lung disease patients. The efficiency of the proposed technique is evaluated using different performance metrics such as Accuracy, Precision, Recall, F-measure, Specificity, confusion matrix, and sensitivity. The high accuracy obtained when comparing the proposed methodology with conventional techniques shows its efficiency of it in improving lung disease diagnosis. Copyright © 2022 Elsevier Ltd

A Framework Automation of COVID-19 Classification through Chest X-rays using Deep Learning over Cloud

There is a huge the spread of Covid-19 pandemic (Corona) in large areas of the country, including modern and rural areas, and due to the scarcity of medical tools and supplies, especially in rural areas. Therefore, artificial intelligence researchers are using technologies to help detect disease early by using chest X-rays to classify whether or not the disease is present. Note that doctors have agreed in more than one scientific article that the initial examination to detect this disease is carried out through chest x-rays, the devices of which are available in most places.Because the Internet is available in most rural areas and in order to reduce the spread of this pandemic, in this paper we built a project by deep transfer learning using an application in Keras called "InceptionV3" on cloud, this model trained and tested 10 thousand images of people with the disease and others where the data distribution was equal to avoid From imbalanced data, and this model will be used across the cloud by web framework so that we can get proactive decisions and avoid spread. This model has been applied in the Department of Respiratory Medicine at Dr. ShankarraoChavan Government Hospital, Nanded, under the supervision of a medical staff headed by Dr. V. R. Kapse, associate professor and head of the department of pulmonary, we have obtained results after training and evaluating the model are training accuracy 97.6%, testing accuracy 97.5%, precision 97.8%, sensitivity 100% and specificity 99.9%. Copyright © 2022, Anka Publishers. All rights reserved.

Deeplearningapproachfor detection of multiple respiratory diseasesfrom chest X-rayanalysis:Asurvey

Deep learning approach for detecting various respiratory diseases hasbeen challenging and mostdemanding research area. Withrapidly increase in number of patients suffering from respiratory diseases quick method hasbecome necessary for classification and detection of respiratory diseases. This survey paper offers a comparative study of various deep learning techniques that can use chest X-raysfordetection of various thoracic diseases.There is possibility of severe respiratory failure in some thoracic diseases if they are not treated in initial stages. Many digital image processing techniques,machine learning and deep learning models have been developed for this purpose[17]. Different forms of existing deep learning techniques including convolutional neural network (CNN), visual geometry group based neural network (VGG-16 and VGG-19) have been developed for respiratory disease prediction. But these all models have some limitations that they do not cover all respiratory diseases including Covid-19, Viral pneumonia and Tuberoculosis on single platform. Therefore, we propose our customized new deep learning model Clx-Net by using data augmentation technique to enlarge the area of available dataset[1][2] to make model more efficient with less time consumption per epoch and provide localization to identify infected region by examining chest X-ray images. Our focus is to develop a new unique deep learning based model Clx-Net which will be able to detect almost all major respiratorydiseases including Covid-19. It will simplify the detection of respiratory diseases and also find the location of infected chest area to make task easy for radiologists.

A Systematic Review of Virtual Reality-Based Education Research Using Latent Dirichlet Allocation: Focus on Topic Modeling Technique

Using virtual reality in the context of education is becoming important since this technology enhances learners' motivation and performance with transfer of learning, problem-solving skills, educational equity, and multisensory learning. Along with the 4th Industrial Revolution and COVID-19, the attention on virtual reality has been increased again. Some studies reviewed the trends of research on virtual reality-based education, mostly focusing on specific subject matters or areas. The purpose of the study is to investigate the change of research trends in the virtual reality-based education field by implementing the topic modeling analysis based on latent Dirichlet allocation (LDA) technique with 6,755 articles published in the last 30 years (between 1992 and 2022). As a result of this study, it was revealed that the research on virtual reality-based education was clearly divided into the following four periods;1992-2011, 2012-2016, 2017-2019, and 2020-2022. The main topics for each period were suggested. Here are three major findings of this study. First, it is identified that the weight of "virtual reality for learning and teaching"has increased in recent years. Second, conversely, it is identified that the proportion of "virtual reality in medical education"has decreased relatively recently. It means it is possible to interpret that the proportion of other topics has relatively increased. Third, the topics of "virtual reality education platform"and "virtual reality-based education in rehabilitation"continue to maintain a certain weight. Limitations of this study and further research suggestions are provided based on the results of the study. © 2022 Daeseok Kim and Tami Im.

Masked Face Detection And Identification By Using Deep Learning Technology

In the wake of the global health disaster brought on by the globally circulating COVID-19 coronavirus. It is currently a research topic in many fields, especially those interesting such as artificial intelligence and new information technologies. Many regulatory agencies now require wearing face masks, particularly in crowded areas involving regular and large-scale human interaction, like inside overcrowded transit facilities, where everyone must wear masks. It is challenging to identify the identity of a person using conventional facial recognition techniques, so it needs developed technology with high accuracy. The paper presents a new system by utilizing the advanced MobileNetV2 network to recognize the person's identity without the need to take off the face mask. The proposed system has trained by using different eight classes of regular people's faces (without wearing face masks) under diverse environmental conditions. The performance of the proposed system demonstrated high efficiency in identifying the identity of the person accurately up to 100%. The recognition process was achieved using Keras with TensorFlow in terms of accuracy and detection speed.

mRNA Vaccine Technology Transfer and Training Hub in South Africa: A partnership model for sustainable vaccine production

Introduction/ Background: While over 6 billion doses of Covid19 vaccines have been administered globally, only 2% of people in Africa have been vaccinated. This uncomfortable reality lead to the establishment of a mRNA vaccine technology transfer and training Hub in South Africa under the COVAX initiative lead by WHO and MPP. Methods: Afrigen Biologics is establishing a technology transfer and training hub for COVID-19 mRNA-based vaccines. In the absence of a technology transfer agreement with the holders of mRNA vaccine technology, Afrigen and its University partners are developing a firstgeneration vaccine, a fast follow-on of mRNA-1273 (Moderna vaccine). The Afrigen-based vaccine technology transfer program will provide sufficient transfer of know-how to allow a competent tech transfer recipient manufacturer in Africa and other LIMCs to successfully manufacture and release mRNA vaccines at scale to support clinical development, national/regional marketing authorization and WHO prequalification, and sustainable supply to meet local and regional vaccine demand. Results: The mRNA Hub at Afrigen has reached key milestones in terms of completion of the facility and start up phase of equipment supply, training of Afrigen core staff in drug substance and drug product production at lab scale. The development of a stable genetic construct that allows transcription of an mRNA molecule at bench scale as well as the encapsulation in a lipid nano particle is underway. This presentation will provide an overview of the progress of the mRNA Hub, its workplan as well as the long-term research and development program and the partnerships supporting the Hub. Impact: The mRNA vaccine technology transfer Hub has created a public private partnership model for sustainable vaccine manufacturing on the African continent. Supported and enabled by the African Union and the African CDC, the mRNA Hub is well positioned to become one of the pillars of the African vaccine manufacturing strategy. Conclusion: The Covid19 pandemic has unleashed significant energy to ensure that Africa implement programs that will ensure sustainable supply of vaccines and preparedness for future pandemics. The Hub and Spoke model is one of the interventions that has the potential to create local innovation and contribute to the supply of vaccines.

Deep Convlutional models on medical data for detection of Covid Lung X-Rays throughTransfer Learning

Lung cancer patients have a greater frequency of COVID-19 infection, pulmonary issues, and worse survival rates as compared to the general population. The world's major professional organisations released new recommendations for the diagnosis, treatment, and follow-up of lung cancer patients as a guide for prioritising cancer care issues during the epidemic. In the modern world, we are battling COVID-2019, a coronavirus-driven pandemic that is among the worst in human history. If the infection is discovered early, the patient can receive treatment right away (before it enters the lower respiratory tract).once the infection has reached the lungs, to look for ground-glass opacity on the chest X-ray caused by fibrosis. Based on the significant differences in the X-ray images of an infected and non-infected person, artificial intelligence systems can be used to determine the presence and severity of illness. In order to extract the features I needed for this study, I used feature extraction from transfer learning, which involves importing a pre-trained CNN model such as the Distributed Deep Convolutional Inception model, Distributed Deep Convolutional VGGNet model, or Distributed Deep Convolutional with ResNet Model and altering the final layer to suit my needs.The model can achieve an F1-score of 0.88 using Distributed Deep Convolutinal VGGNet, the highest of all pretrained models. Additionally, the COVID-19- related X-rays are broken down into three severity levels: moderate, medium, and severe. The data are analysed using the F1-Score because precision and recall are both crucial elements in this investigation. The confusion matrix and the results for the F1-Score, Precision, Recall, and overall Accuracy are also supplied to give a full analysis of the Model performance. The proposed strategies have had a significant impact on the nation as a warning to society.

Deep Convlutional models on medical data for detection of Covid Lung X-Rays through Transfer Learning

Lung cancer patients have a greater frequency of COVID-19 infection, pulmonary issues, and worse survival rates as compared to the general population. The world's major professional organisations released new recommendations for the diagnosis, treatment, and follow-up of lung cancer patients as a guide for prioritising cancer care issues during the epidemic. In the modern world, we are battling COVID-2019, a coronavirus-driven pandemic that is among the worst in human history. If the infection is discovered early, the patient can receive treatment right away (before it enters the lower respiratory tract).once the infection has reached the lungs, to look for ground-glass opacity on the chest X-ray caused by fibrosis. Based on the significant differences in the X-ray images of an infected and non-infected person, artificial intelligence systems can be used to determine the presence and severity of illness. In order to extract the features I needed for this study, I used feature extraction from transfer learning, which involves importing a pre-trained CNN model such as the Distributed Deep Convolutional Inception model, Distributed Deep Convolutional VGGNet model, or Distributed Deep Convolutional with ResNet Model and altering the final layer to suit my needs.The model can achieve an F1-score of 0.88 using Distributed Deep Convolutinal VGGNet, the highest of all pretrained models. Additionally, the COVID-19-related X-rays are broken down into three severity levels: moderate, medium, and severe. The data are analysed using the F1-Score because precision and recall are both crucial elements in this investigation. The confusion matrix and the results for the F1-Score, Precision, Recall, and overall Accuracy are also supplied to give a full analysis of the Model performance. The proposed strategies have had a significant impact on the nation as a warning to society.

Designing a Deep Learning Hybrid Using CNN and Inception V3 Transfer Learning to Detect the Aggression Level of Deep Obsessive Compulsive Disorder in Children

The usage of Artificial intelligence in medical arena has proved to be a game changer in the detection and diagnosis of several medical conditions. In the current digital era, children with stressful medical issues are suffering from Deep Obsessive-Compulsive Disorder (DOCD). This kind of mental stress occurs in children because of the continuous usage of gadgets such as mobile phone, playing games using play stations, watching videos on tablets, etc. In most of the possibilities, single children are the ones affected with several obsessions such as stubborn activities, fighting for selfish priorities and so on. In medical terms, these kinds of complex behavioral changes are identified as DOCD. Genetic behaviors sometimes in a few group of children are also noticed as a modality difference. As symptoms are psychiatric impairment, such a child remains isolated, abnormal silence, being obsessive and repeating irrelevant words, high stress or anxiety. All medical challenges could be treated as healthcare research metrics and the gradual increase in DOCD disorder among children of this generation can be considered too. Early detection of DOCD is essential as it can help in early diagnosis but techniques to do so is unavailable currently. Deep learning-an artificial intelligence method can be utilized to detect DOCD, diagnose and treat it and bring about a positive character in children. Behavior changes in children can be classified and detected using transfer learning algorithms. In COVID-19 pandemic situation, 3% of DOCD has increased to 10-15% as a disorder. This information is retrieved from children by monitoring negative activities, unusual behavior such as nail biting, removing spectacles and placing them in the wrong place, watching tablets, mobile phones and television for more hours. Using Convolutional Neural Networks (CNN), input such as MRI (Magnetic resonance Imaging) is used for experimenting the variations in behavior with the high dimension that are analyzed from the image dataset. Using Transfer Learning with Inception V3-, CNN generalization of misophonia level can be statistically analyzed to avoid overfitting problems. By employing AI techniques, the aggression level can be predicted using data augmentation method with better accuracy and a low error rate than the existing systems. In the research it is observed that using the model employing Inception-V3 transfer learning CNN a better prediction of aggression levels can be achieved in comparison to the existing CNN model used.

Decision support system in radiology for fast diagnostics of thoracic diseases under COVID-19 pandemic conditions

In the present article the relevance of using DSS under the current conditions for image recognition and, as a more specific application, for the purpose of additional assistance rendered to medical experts (radiologists) in their decision-making and preparing findings upon assessment of X-ray images is considered. The paper analyzes the requirements for some expert DSS and their main characteristics that they should have;considered and selected is the necessary software for making rapid diagnoses of diseases of the thorax. All these modern requirements and characteristics are met by the Deep Learning Studio (DLS) software, which allows using deep convolutional neural network Inception V3 to teach this network and further obtain optimal results in the recognition and diagnosis of diseases of the thorax by assessing X-ray images. As a result of this study, a ready-made DSS intended for use by medical institutions for additional assistance to radiologists to prepare findings according to X-ray images has been obtained.

EpistoNet: An ensemble of deep convolutional neural networks using mixture of discriminative experts for detecting COVID-19 on chest X-ray images

Introduction: The Coronavirus has spread across the globe and infected millions of people, having devastating effect on the global public health and economies. A fast diagnostic system should be implemented to mitigate the impact of the virus and save lives. In this study, we propose a decision tree-based ensemble model using two mixtures of discriminative experts (MoE) to classify COVID-19 and non-COVID-19 lung infections on chest X-ray images. The Epistocracy algorithm, a hyper-heuristic evolutionary method, is employed to optimize the neural networks used in this work. Using this approach can help detect COVID-19 cases and accelerate treatment of those who need it the most. Data: we collected 2,500 chest X-ray images from Henry Ford Health System consisting of 1,250 Covid images and 1,250 non-Covid images. The input images have been cropped and resized to 224 by 224 pixels. Out of 2,500 images, we left out 500 images containing 250 Covid and 250 non-Covid for testing. The rest, 2,000 images, were used 80% for training and 20% for validation. Methods and Results: To improve the accuracy of the proposed model, first we divided our 2,000 images into 5 different clusters using K-Means clustering algorithm with VGG16 feature extractor to help build strong discriminative expert models to be used in our proposed classifier. We trained VGG16, VGG19, InceptionV3, InceptionResNetV2, MobileNetV2, EfficientNetB7, Xception, and DenseNet201 to classify each cluster into Covid and non-Covid cases. The best result was obtained from VGG16 as a base model with a deep neural network as a head model optimized by Epistocracy algorithm. Then we built a mixture of transfer learning-based experts consisting of 5 different VGG16 models supervised by InceptionV3 as a gating network. Finally, we built a decision tree-based ensemble model to determine the classification of the data using two different MoEs with highest accuracies. As a result, for initial clusters c1, c2, c3, c4, and c5 we obtained validation accuracy of 92.50%, 86.30%, 86.51%, 85.34%, and 93.62% respectively. The first MoE had 93.75% accuracy on validation, and the second MoE had 94.25%. The final ensemble model on average obtained 94% accuracy on the testing dataset. More specifically, we got 96% accuracy on Covid images and 92% accuracy on non-Covid. Conclusion: we showed that an ensemble model consisting of two mixtures of cluster-based discriminative convolutional neural network experts can be used to detect Covid from non-Covid with high accuracy, and Epistocracy algorithm can be effectively used to optimize the hyper-parameters of the proposed models.

When knowing is not enough: A narrative exploration of how K-12 teachers make decisions about the transfer of critical competencies from professional learning to daily practice

School districts spend millions of dollars each year to provide training and learning to staff working in direct and indirect service to students (National Council on Teacher Quality, 2021). This financial commitment says nothing about what is even more important: the need for school employees and the systems in which we work to serve students more effectively. Despite vast allocations of time and money and presumably best intentions for better social and academic outcomes for students, very little data exist that reflect regular transfer and application of training/learning into professional practice (Nittler et al., 2015). By and large, schools and school systems look the same today as they did 50+ years ago despite the fact that the world looks very different and so much more is known about the cognitive process and contextual contributors involved in erudition development. Teacher application of critical competencies such as cultural responsiveness, trauma informed practices, social emotional learning and basic neuroscience in the ways they conceptualize and implement instructional practices may not be easily apparent during casual observation, yet they are inextricably linked to positive academic and social outcomes for students, thus imperative to effective professional practice. This study investigates the ways in which professional educators make decisions about the transfer and application of professional learning centered on critical competencies (soft skills) in their daily work. Narrative Inquiry (NI) provided the methodological frame for this exploratory study that through thematic analysis surfaced five key factors influencing learning transfer: Instructor/Presenter/Facilitator;Connection to Lived Experience;Relevance to Job Assignment;Alignment with Self-Identity;and COVID-19. This dissertation is available in open access at AURA (https://aura.antioch.edu ) and OhioLINK ETD Center (https://etd.ohiolink.edu). (PsycInfo Database Record (c) 2022 APA, all rights reserved)

Transfer Learning-Based Attention Gated Siamese Network for Human and SARS-CoV-2 Protein Interactions

For the past year SARS-CoV-2 has affected the lives of people around the globe. Therefore, research community is continuously putting in their best efforts to find a solution to curb and cure the disease. SARS-CoV-2 is a 29.9k bp long sequence genome comprising of 25 different proteins among which spike glycoprotein plays a vital role in interaction with the host cells. Hence, majority of the scientific studies were focused towards targeting the spike region for the vaccine design against the contagious virus. Thorough study of protein-protein interaction between human and virus can help us in better understanding and management of this disease. For this purpose, an Attention gated Siamese framework is utilized from which a consensus of prominent features and contextual information is taken into account to identify the influence of protein sequences. Moreover, to obtain the pattern of interacting pairs of human and SARS-CoV-2 proteins, a transfer learning-based approach is opted from the proposed network through which we obtained an accuracy of 85%. Additionally, by using this model, we identified that there were 30, 13 and 17 human proteins interacting with spike glycoprotein, nucleocapsid and membrane respectively, having predictive interaction of above 90% for each of the interactions.

Machine learning screening of COVID-19 patients based on X-ray images for unbalanced classes

Background: COVID-19 is a pandemic that had already infected more than forty-six million people and caused more than a million deaths by 1st of November 2020. The virus pandemic appears to have had a catastrophic effect on the global population's safety. Therefore, efficient detection of infected patients is a key phase in the battle against COVID-19. One of the main screening methods is radiological testing. The goal of this study is using chest X-ray images to detect COVID-19 pneumonia patients while optimizing detection efficiency. Methods: As shown in Figure 1, we combined three methods to detect COVID-19 namely: convolutional neural network, transfer learning, and the focal loss1 function which are used for unbalanced classes, to build three binary classifiers which are COVID-19 versus normal, COVID-19 versus pneumonia, and COVID-19 versus normal pneumonia (normal and pneumonia). The database used2 includes a mixture of 400 COVID-19, 1,340 viral pneumonia, 2,560 bacterial pneumonia, and 1,340 normal chest X-ray images for training, validation, and testing of four pre-trained deep convolutional neural networks. Then, the pre-trained model that gives the best results was chosen to improve its performances by two enhancement techniques which are image augmentation, allowing us to reach approximately 2,500 images per class, and the adjustment of focal loss hyperparameters. Results: A comparative study was conducted of our proposed classifiers with well-known classifiers and obtained much better results in terms of accuracy, specificity, sensitivity and precision, as illustrated in Table 1. Conclusion: The high performance of this computer-aided diagnostic technique may greatly increase the screening speed and reliability of COVID-19 diagnostic cases. Particularly, at the crowded emergency services, it will be particularly helpful in this pandemic when the risk of infection and the necessity for prevention initiatives run contrary to the available resources.