Fast automated detection of COVID-19 from medical images using convolutional neural networks

Abstract Coronavirus disease 2019 (COVID-19) is a global pandemic posing significant health risks. The diagnostic test sensitivity of COVID-19 limited due to irregularities in specimen handling. We propose deep learning framework that identifies from medical images as an auxiliary testing method improve sensitivity. use pseudo-coloring methods and platform for annotating X-ray computed tomography train the convolutional neural network, which achieves performance similar experts provides high scores multiple statistical indices (F1 > 96.72% (0.9307, 0.9890) specificity >99.33% (0.9792, 1.0000)). Heatmaps are used visualize salient features extracted by network. network-based regression strong correlations between lesion areas five clinical indicators, resulting accuracy classification framework. proposed represents potential computer-aided diagnosis practice.