COVID-19 detection in X-ray images using convolutional neural networks

Machine Learning with Applications

Jesús Alejandro Alzate-Grisales¹  Mario Alejandro Bravo-Ortiz²  Harold Brayan Arteaga-Arteaga³  Daniel Arias-Garzón⁴  Jose Manuel Saborit-Torres⁴  Joaquim Ángel Montell Serrano⁴  Alejandro Mora-Rubio⁴  Simon Orozco-Arias⁴  Oscar Cardona-Morales⁵  Maria de la Iglesia Vayá⁵  Reinel Tabares-Soto⁵ 

[1] Corresponding authors.;Department of Systems and Informatics, Universidad de Caldas, Manizales 170004, Colombia;Department of Computer Science, Universidad Autonóma de Manizales, Manizales 170001, Colombia;Department of Electronics and Industrial Automation, Universidad Autonóma de Manizales, Manizales 170001, Colombia;Unidad Mixta de Imagen Biomédica FISABIO-CIPF. Fundación para el Fomento de la Investigación Sanitario y Biomédica de la Comunidad Valenciana, Valencia 46020, Spain;

关键词: COVID-19;    Deep learning;    Transfer learning;    X-ray;    Segmentation;   

DOI  :  

来源: DOAJ

【 摘 要 】

COVID-19 global pandemic affects health care and lifestyle worldwide, and its early detection is critical to control cases’ spreading and mortality. The actual leader diagnosis test is the Reverse transcription Polymerase chain reaction (RT-PCR), result times and cost of these tests are high, so other fast and accessible diagnostic tools are needed. Inspired by recent research that correlates the presence of COVID-19 to findings in Chest X-ray images, this papers’ approach uses existing deep learning models (VGG19 and U-Net) to process these images and classify them as positive or negative for COVID-19. The proposed system involves a preprocessing stage with lung segmentation, removing the surroundings which does not offer relevant information for the task and may produce biased results; after this initial stage comes the classification model trained under the transfer learning scheme; and finally, results analysis and interpretation via heat maps visualization. The best models achieved a detection accuracy of COVID-19 around 97%.

【 授权许可】

Unknown