X-Ray Image Classification for COVID-19 Using Transfer Learning

SARS-CoV-2, the cause of one of the significant pandemics in history, first appeared in Wuhan, China. It spreads rapidly, with symptoms like fever, cough, tiredness, and loss of taste or smell. We came up with many measures where the most effective was vaccines. Yet it's not enough against the rapidly appearing waves of SARS-CoV-2. A deep learning algorithm has proven efficient in detecting Covid-19 based on pneumonia and respiratory problems. These problems have been identified with the help of CT scans and X-ray images. It'll make it a lot easier to determine who's Infected and would save a lot of time and expenses overall would provide for extensive relief in the Covid-19 pandemic. This paper uses publically available COVID-19 X-Ray and CT Scan images to create a dataset. The Deep Learning based model is used to train and test the dataset. In the experiment, the overall accuracy is 98%, and in the testing process, the overall accuracy is 99%. © 2023 The authors and IOS Press.

COVID-19: X-ray image classification using transfer learning DCNN approach

COVID-19 is a coronavirus that causes sickness in the human respiratory system. It is the most recent virus that is wreaking havoc on the entire world. It spreads mainly through contact with an infected person. There are some vaccinations available to prevent this condition now. The flu causes symptoms such as fever, coughing, and breathing difficulties in humans. COVID-19: Classification of X-Ray Images This paper suggests using a Deep Convolution Neural Network-based Transfer Learning methodology. Deep CNN learns picture patterns and classifies X-RAY pictures using transfer learning technology. A dataset is created using publicly available photos of COVID-19 X-Ray. All images have been resized and rotated by 2 to 20 degrees. The file contains 6677 COVID-19 pictures and 5753 stock pictures. DCNN predictability is 99.64 percent on a training set, while on a test set, it is 99.79 percent. After the transfer of learning, predictive accuracy on the training set is 99.19 percent, while predictive accuracy on the test set is 99.31 percent. © 2022 Author(s).

COVID-19 PREDICTION USING TRANSFER-LEARNING ON RT-PCR CONFIRMED CXR-IMAGES

Corona-virus is a disease which caused immense destruction to human lives in 21st century. This virus outbreak is considered as an epidemic that spread globally. Crores of people are infected by this virus all over the world. Early detection of the virus is very much important to overcome Covid-19 crisis. This model proposes a convolution neural network model implemented using VGG-19accompanied with Transfer LearningTechnique for the Covid-19 Detection. The Covid-19 dataset considered in this model is a verified report of positive cases confirmed by both RT-PCR and CXR images. Initially, One Hot Encoding Method is used for CXR image data conversion and then pre-processing is done to extract features and then filtered data is forwarded through the VGG-19 and is further processed to Fully Connected Layers. Therefore, the model is later fine-tuned to achieve better classification results. The achieved model accuracy is around 0.94 with a loss is about 0.55. © 2022, Engg Journals Publications. All rights reserved.

Viability and Applicability of Deep Learning Approach for COVID-19 Preventive Measures Implementation

The deadliest COVID-19 (SARS-CoV-2) is expanding steadily and internationally due to which the nation economy almost come to a complete halt;citizens are locked up;activity is stagnant and this turn toward fear of government for the health predicament. Public healthcare organizations are mostly in despair need of decision-making emerging technologies to confront this virus and enable individuals to get quick and efficient feedback in real-time to prevent it from spreading. Therefore, it becomes necessary to establish auto-mechanisms as a preventative measure to protect humanity from SARS-CoV-2. Intelligence automation tools as well as techniques could indeed encourage educators and the medical community to understand dangerous COVID-19 and speed up treatment investigations by assessing huge amounts of research data quickly. The outcome of preventing approach has been used to help evaluate, measure, predict, and track current infected patients and potentially upcoming patients. In this work, we proposed two deep learning models to integrate and introduce the preventive sensible measures like face mask detection and image-based X-rays scanning for COVID-19 detection. Initially, face mask detection classifier is implemented using VGG19 which identifies those who did not wear a face mask in the whole crowd and obtained 99.26% accuracy with log loss score 0.04. Furthermore, COVID-19 detection technique is applied onto the X-ray images that used a Xception deep learning model which classifies whether such an individual is an ordinary patient or infected from COVID-19 and accomplished overall 91.83% accuracy with 0.00 log loss score. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Federated Transfer Learning-Based IDS for the Internet of Medical Things (IoMT)

The Internet of Medical Things (IoMT) is a set of medical devices and applications that connect to healthcare systems through the Internet. Those devices are equipped with communication technologies that allow them to communicate with each other and the Internet. Reliance on the IoMT is increasing with the increase in epidemics and chronic diseases such as COVID-19 and diabetes;with the increase in the number of IoMT users and the need for electronic data sharing and virtual services, cyberattacks in the healthcare sector for accessing confidential patient data has been increasing in the recent years. The healthcare applications and their infrastructures have special requirements for handling sensitive users' data and the need for high availability. Therefore, securing healthcare applications and data has attracted special attention from both industry and researchers. In this paper, we propose a Federated Transfer Learning-based Intrusion Detection System (IDS) to secure the patient's healthcare-connected devices. The model uses Deep Neural Network (DNN) algorithm for training the network and transferring the knowledge from the connected edge models to build an aggregated global model and customizing it for each one of the connected edge devices without exposing data privacy. CICIDS2017 dataset has been used to evaluate the performance in terms of accuracy, detection rate, and average training time. In addition to preserving data privacy of edge devices and achieving better performance, our comparison indicates that the proposed model can be generalized better and learns incrementally compared to other baseline ML/DL algorithms used in the traditional centralized learning schemes. © 2021 IEEE.

A complete framework for accurate recognition and prognosis of COVID-19 patients based on deep transfer learning and feature classification approach.

The sudden appearance of COVID-19 has put the world in a serious situation. Due to the rapid spread of the virus and the increase in the number of infected patients and deaths, COVID-19 was declared a pandemic. This pandemic has its destructive effect not only on humans but also on the economy. Despite the development and availability of different vaccines for COVID-19, scientists still warn the citizens of new severe waves of the virus, and as a result, fast diagnosis of COVID-19 is a critical issue. Chest imaging proved to be a powerful tool in the early detection of COVID-19. This study introduces an entire framework for the early detection and early prognosis of COVID-19 severity in the diagnosed patients using laboratory test results. It consists of two phases (1) Early Diagnostic Phase (EDP) and (2) Early Prognostic Phase (EPP). In EDP, COVID-19 patients are diagnosed using CT chest images. In the current study, 5, 159 COVID-19 and 10, 376 normal computed tomography (CT) images of Egyptians were used as a dataset to train 7 different convolutional neural networks using transfer learning. Data augmentation normal techniques and generative adversarial networks (GANs), CycleGAN and CCGAN, were used to increase the images in the dataset to avoid overfitting issues. 28 experiments were applied and multiple performance metrics were captured. Classification with no augmentation yielded 99.61 % accuracy by EfficientNetB7 architecture. By applying CycleGAN and CC-GAN Augmentation, the maximum reported accuracies were 99.57 % and 99.14 % by MobileNetV1 and VGG-16 architectures respectively. In EPP, the prognosis of the severity of COVID-19 in patients is early determined using laboratory test results. In this study, 25 different classification techniques were applied and from the different results, the highest accuracies were 98.70 % and 97.40 % reported by the Ensemble Bagged Trees and Tree (Fine, Medium, and Coarse) techniques respectively.

CovH2SD: A COVID-19 detection approach based on Harris Hawks Optimization and stacked deep learning.

Starting from Wuhan in China at the end of 2019, coronavirus disease (COVID-19) has propagated fast all over the world, affecting the lives of billions of people and increasing the mortality rate worldwide in few months. The golden treatment against the invasive spread of COVID-19 is done by identifying and isolating the infected patients, and as a result, fast diagnosis of COVID-19 is a critical issue. The common laboratory test for confirming the infection of COVID-19 is Reverse Transcription Polymerase Chain Reaction (RT-PCR). However, these tests suffer from some problems in time, accuracy, and availability. Chest images have proven to be a powerful tool in the early detection of COVID-19. In the current study, a hybrid learning and optimization approach named CovH2SD is proposed for the COVID-19 detection from the Chest Computed Tomography (CT) images. CovH2SD uses deep learning and pre-trained models to extract the features from the CT images and learn from them. It uses Harris Hawks Optimization (HHO) algorithm to optimize the hyperparameters. Transfer learning is applied using nine pre-trained convolutional neural networks (i.e. ResNet50, ResNet101, VGG16, VGG19, Xception, MobileNetV1, MobileNetV2, DenseNet121, and DenseNet169). Fast Classification Stage (FCS) and Compact Stacking Stage (CSS) are suggested to stack the best models into a single one. Nine experiments are applied and results are reported based on the Loss, Accuracy, Precision, Recall, F1-Score, and Area Under Curve (AUC) performance metrics. The comparison between combinations is applied using the Weighted Sum Method (WSM). Six experiments report a WSM value above 96.5%. The top WSM and accuracy reported values are 99.31% and 99.33% respectively which are higher than the eleven compared state-of-the-art studies.

Hybrid COVID-19 segmentation and recognition framework (HMB-HCF) using deep learning and genetic algorithms.

COVID-19 (Coronavirus) went through a rapid escalation until it became a pandemic disease. The normal and manual medical infection discovery may take few days and therefore computer science engineers can share in the development of the automatic diagnosis for fast detection of that disease. The study suggests a hybrid COVID-19 framework (named HMB-HCF) based on deep learning (DL), genetic algorithm (GA), weighted sum (WS), and majority voting principles in nine phases. Its segmentation phase suggests a lung segmentation algorithm using X-Ray images (named HMB-LSAXI) for extracting lungs. Its classification phase is built from a hybrid convolutional neural network (CNN) architecture using an abstractly-designed CNN (named HMB1-COVID19) and transfer learning (TL) pre-trained models (VGG16, VGG19, ResNet50, ResNet101, Xception, DenseNet121, DenseNet169, MobileNet, and MobileNetV2). The hybrid CNN architecture is used for learning, classification, and parameters optimization while GA is used to optimize the hyperparameters. This hybrid working mechanism is combined in an overall algorithm named HMB-DLGA. The study experiments implemented the WS approach to evaluate the models' performance using the loss, accuracy, F1-score, precision, recall, and area under curve (AUC) metrics with different pre-defined ratios. A collected, combined, and unified X-Ray dataset from 8 different public datasets was used alongside the regularization, dropout, and data augmentation techniques to limit the overall overfitting. The applied experiments reported state-of-the-art metrics. VGG16 reported 100% WS metric (i.e., 0.0097, 99.78%, 0.9984, 99.89%, 99.78%, and 0.9996 for the loss, accuracy, F1, precision, recall, and AUC respectively) concerning the highest WS. It also reported a 99.92% WS metric (i.e., 0.0099, 99.84%, 0.9984, 99.84%, 99.84%, and 0.9996 for the loss, accuracy, F1, precision, recall, and AUC respectively) concerning the last reported WS result. HMB-HCF was validated on 13 different public datasets to verify its generalization. The best-achieved metrics were compared with 13 related studies. These extensive experiments' target was the applicability verification and generalization.