Performance Analysis of Convolutional Neural Network Architectures for the Identification of COVID-19 from Chest X-ray Images

The 2019 Novel Coronavirus (COVID-19) has spread quickly over the world and continues to impact the health and well-being of people. The application of deep learning coupled with radiological images is effective for early diagnosis and prevention of the spread. In this study, we introduced a 2D Convolutional Neural Network (CNN) to automatically diagnose Chest X-ray images for multi-class classification (COVID-19 vs. Viral Pneumonia vs. Normal). The objective of the research is to maximize the accuracy of detection by altering various internal parameters of a 2D CNN architecture. A dataset consisting of 1000 COVID-19, 1000 Viral Pneumonia, and 1000 Normal images was considered, and preprocessing steps and augmentation strategies were applied. The training and evaluation of the results were performed on eight 2D CNN architectures with internal parameters changed specifically in each case, and a COVID-19 classification model was proposed. Our proposed computer-aided diagnostic tool produced a significant performance with a classification accuracy of 97.3 %, a sensitivity of 97.3 %, specificity of 98.7%, and precision of 97.4 % on test datasets. These results suggest that it can reliably detect COVID-19 cases and expedite treatment to those in the most need. © 2022 IEEE.

On COVID-19 Prediction Using Asynchronous Federated Learning-Based Agile Radiograph Screening Booths

To combat the novel coronavirus (COVID-19) spread, the adoption of technologies including the Internet of Things (IoT) and deep learning is on the rise. However, the seamless integration of IoT devices and deep learning models for radiograph detection to identify the presence of glass opacities and other features in the lung is yet to be envisioned. Moreover, the privacy issue of the collected radiograph data and other health data of the patients has also arisen much concern. To address these challenges, in this paper, we envision a federated learning model for COVID-19 prediction from radiograph images acquired by an X-ray device within a mobile and deployable screening resource booth node (RBN). Our envisioned model permits the privacy-preservation of the acquired radiograph by performing localized learning. We further customize the proposed federated learning model by asynchronously updating the shallow and deep model parameters so that precious communication bandwidth can be spared. Based on a real dataset, the effectiveness of our envisioned approach is demonstrated and compared with baseline methods.

A Comparative Study of CNN Transfer Learning Classification Algorithms with Segmentation for COVID-19 Detection from CT Scan Images

After it's inception, COVID-19 has spread rapidly all across the globe. Considering this outbreak, by far, it is the most decisive task to detect early and isolate the patients quickly to contain the spread of this virus. In such cases, artificial intelligence and machine learning or deep learning methods can come to aid. For that purpose, we have conducted a qualitative investigation to inspect 12 off-the-shelf Convolution Neural Network (CNN) architectures in classifying COVID-19 from CT scan images. Furthermore, a segmentation algorithm for biomedical images - U-Net, is analyzed to evaluate the performance of the CNN models. A publicly available dataset (SARS-COV-2 CT-Scan) containing a total of 2481 CT scan images is employed for the performance evaluation. In terms of feature extraction by excluding the segmentation technique, a performance of 88.60% as the F1 Score and 89.31% as accuracy is achieved by training DenseNet169 architecture. Adopting the U-Net segmentation method, we accomplished the most optimal accuracy and F1 Scores as 89.92% and 89.67% respectively on DenseNet201 model. Furthermore, evaluating the performances, we can affirm that a combination of a Transfer Learning architecture with a segmentation technique (U-Net) enhances the performance of the classification model.

Potential of betacyanin as inhibitor of SARS-CoV-2 revealed by molecular docking study

Covid-19 is a global pandemic where an effective drug has yet to be found. A new coronavirus species, SARS-CoV-2 causes this disease. Several studies have been conducted on medicinal plant-based lead compounds to find antidotes for this virus. One of the fruits that with a high betacyanin content is super red dragon fruit produced by plant Hylocereus costaricensis. Betacyanin, besides having anti-inflammatory and immunomodulatory activities, also has antiviral activity. Therefore, this study aimed to evaluate betacyanin's interaction with several SARS-CoV-2 receptors by observing its binding affinity and compared it with the nelfinavir and hydroxychloroquine sulfate that have been recommended for treating COVID-19. This research was an in silico study using computer software to assess binding affinity simulations based on molecular docking. The results of this study indicated that betacyanin had a good affinity with several receptors so that it has the potential to be developed as a lead compound to overcome COVID-19. Based on its binding affinity value, betacyanin's ability was comparable to nelfinavir and hydroxychloroquine sulfate recommended by WHO as a therapeutic agent for COVID-19. © Published under licence by IOP Publishing Ltd.

Bilateral adrenal hemorrhage in a man with severe COVID-19 pneumonia.

This case report demonstrates bilateral adrenal hemorrhage in a fifty-year old man with severe COVID-19 pneumonia. We discuss how adrenal hemorrhage can be one of the possible complications of COVID-19. The case also shows how adrenal hemorrhage can be diagnosed incidentally in a scan performed for a different reason given the difficulty of clinical diagnosis and the non-specific clinical presentation.

Automatic Detection of COVID-19 Disease in Chest X-Ray Images using Deep Neural Networks

The worldwide spread of COVID-19 has marked a devastating impact on the global economy and public health. One of the significant steps of COVID-19 affected patient's treatment is the faster and accurate detection of the symptoms which is the motivational center of this study. In this paper, we have analyzed the performances of six artificial deep neural networks (2-D CNN, ResNet-50, InceptionResNetV2, InceptionV3, DenseNet201, and MobileNetV2) for COVID-19 detection from the chest X-rays. Our dataset consists of 2905 chest X-rays of three categories: COVID-19 affected (219 cases), Viral Pneumonia affected (1345 cases), and Normal Chest X-rays (1341 cases). Among the implemented neural networks, ResNet-50 demonstrated reasonable performance in classifying different cases with an overall accuracy of 96.91%. Most importantly, the model has shown a significantly good performance in detecting the COVID-19 cases in the test dataset (Precision = 1.00, Sensitivity = 1.00, Specificity = 1.00, and F1-score = 1.00). Therefore, among the deep neural networks presented in this paper, ResNet-50 can be adapted as a reliable method for faster and accurate COVID-19 affected case detection. © 2020 IEEE.