Search

Computerized disease detection systems (CDDs) have proven effective for automatic screening in recent years. Among the standard procedures in hospitals for faster and more accurate diagnosis is medical imaging-based disease screening. We aim to develop a CDD that detects COVID-19 using chest X-rays pre-trained vision transformers (PVTs). This scheme includes the following steps: (1) collecting images and resizing them, (2) implementing PVT for feature extraction, and (3) binary classifying the results and validating the proposed schemes. To prove the merit of the developed scheme, 4800 images (2400 normal and 2400 COVID-19) are analyzed. MLP classifiers verify the PVT performance using patch sizes of 6, 12, and 24. A patch size 24 results in 97.5% accuracy for the proposed CDD system. When patch sizes are increased to 12, accuracy increases to over 98%. For this specific task, smaller patch sizes are more effective. These high-accuracy results demonstrate the effectiveness of the developed scheme for detecting COVID-19 in chest X-rays. [ABSTRACT FROM AUTHOR]

Computerized disease detection systems (CDDs) have proven effective for automatic screening in recent years. Among the standard procedures in hospitals for faster and more accurate diagnosis is medical imaging-based disease screening. We aim to develop a CDD that detects COVID-19 using chest X-rays pre-trained vision transformers (PVTs). This scheme includes the following steps: (1) collecting images and resizing them, (2) implementing PVT for feature extraction, and (3) binary classifying the results and validating the proposed schemes. To prove the merit of the developed scheme, 4800 images (2400 normal and 2400 COVID-19) are analyzed. MLP classifiers verify the PVT performance using patch sizes of 6, 12, and 24. A patch size 24 results in 97.5% accuracy for the proposed CDD system. When patch sizes are increased to 12, accuracy increases to over 98%. For this specific task, smaller patch sizes are more effective. These high-accuracy results demonstrate the effectiveness of the developed scheme for detecting COVID-19 in chest X-rays. [ABSTRACT FROM AUTHOR]

• An explainable Deep Neural Network to realize online monitoring of COVID-19 gravity by using ECG beat signal. • The network is based on ResNet18 with few enhancements: 1) Adding LSTM Layer inside the residual block, for regenerating the backpropagation error and further extracting the involved time-varying features; 2) Using LeakyReLU instead of ReLU for increasing the performances of the model. • The model outperformed the related existing methods in terms of accuracy and robustness. • The model originally identifies the ST interval of the ECG pattern, as the most prominent key features affected by the virus. Wearable systems measuring human physiological indicators with integrated sensors and supervised learning-based medical image analysis (e.g. ECG, X-ray, CT or ultrasound images for lung or the chest) have been considered relevant tools for COVID-19 monitoring and diagnosis. However, these two technical roadmaps have their respective advantages and drawbacks. The current wearable systems enable to realize real-time monitoring of COVID-19 but are limited to its basic symptoms only, neither allowing to distinguish it from other diseases nor performing deep analysis. Current medical image analysis can provide accurate decision support for diagnosis but rarely deals with real-time data processing. In this context, we propose a new wearable system by combining the advantages of these two technical roadmaps. Considering that electrocardiogram (ECG) has been proved relevant to evolution of COVID-19 symptoms, the proposed wearable system will integrate an explainable Deep Neural Network to realize online monitoring of COVID-19 gravity by using ECG beat signal. This paper will focus on the Deep Neural Network model named X-RCRNet. The network is based on ResNet18 but with few enhancements: 1) LSTM Layers for regenerating the backpropagation error and further extracting the involved time-varying features; 2) LeakyReLU for increasing the performances of the model. With an accuracy of 96.48 % after experiments, our model has not only outperformed the existing methods in terms of accuracy and robustness, but also originally identify the ST interval of the ECG pattern, as the most prominent key features affected by the virus. [ABSTRACT FROM AUTHOR]