The Practicality of Deep Learning Algorithms in COVID-19 Detection: Application to Chest X-ray Images
Open Access
- 13 June 2021
- journal article
- research article
- Published by MDPI AG in Algorithms
- Vol. 14 (6), 183
- https://doi.org/10.3390/a14060183
Abstract
Since January 2020, the outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has affected the whole world, producing a respiratory disease that can become severe and even cause death in certain groups of people. The main method for diagnosing coronavirus disease 2019 (COVID-19) is performing viral tests. However, the kits for carrying out these tests are scarce in certain regions of the world. Lung conditions as perceived in computed tomography and radiography images exhibit a high correlation with the presence of COVID-19 infections. This work attempted to assess the feasibility of using convolutional neural networks for the analysis of pulmonary radiography images to distinguish COVID-19 infections from non-infected cases and other types of viral or bacterial pulmonary conditions. The results obtained indicate that these networks can successfully distinguish the pulmonary radiographies of COVID-19-infected patients from radiographies that exhibit other or no pathology, with a sensitivity of 100% and specificity of 97.6%. This could help future efforts to automate the process of identifying lung radiography images of suspicious cases, thereby supporting medical personnel when many patients need to be rapidly checked. The automated analysis of pulmonary radiography is not intended to be a substitute for formal viral tests or formal diagnosis by a properly trained physician but rather to assist with identification when the need arises.Keywords
Funding Information
- Qassim University (25650)
This publication has 49 references indexed in Scilit:
- Host and infectivity prediction of Wuhan 2019 novel coronavirus using deep learning algorithmPublished by Cold Spring Harbor Laboratory ,2020
- Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCREurosurveillance, 2020
- From local explanations to global understanding with explainable AI for treesNature Machine Intelligence, 2020
- Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based LocalizationInternational Journal of Computer Vision, 2019
- Scalable Digital Neuromorphic Architecture for Large-Scale Biophysically Meaningful Neural Network With Multi-Compartment NeuronsIEEE Transactions on Neural Networks and Learning Systems, 2019
- Explainable machine-learning predictions for the prevention of hypoxaemia during surgeryNature Biomedical Engineering, 2018
- Real-Time Neuromorphic System for Large-Scale Conductance-Based Spiking Neural NetworksIEEE Transactions on Cybernetics, 2018
- Xception: Deep Learning with Depthwise Separable ConvolutionsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2017
- TensorFlow: learning functions at scalePublished by Association for Computing Machinery (ACM) ,2016
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015