Deep Learning Applied to COVID-19 Detection in X-Ray Images

Deep Learning Applied to COVID-19 Detection in X-Ray Images

Harold Brayan Arteaga-Arteaga (Universidad Autónoma de Manizales, Colombia), Melissa delaPava (Universidad Nacional de Colombia, Colombia), Alejandro Mora-Rubio (Universidad Autónoma de Manizales, Colombia), Mario Alejandro Bravo-Ortíz (Universidad Autónoma de Manizales, Colombia), Jesus Alejandro Alzate-Grisales (Universidad Autónoma de Manizales, Colombia), Daniel Arias-Garzón (Universidad Autónoma de Manizales, Colombia), Luis Humberto López-Murillo (Universidad Nacional de Colombia, Colombia), Felipe Buitrago-Carmona (Universidad Autónoma de Manizales, Colombia), Juan Pablo Villa-Pulgarín (Universidad Autónoma de Manizales, Colombia), Esteban Mercado-Ruiz (Universidad Autónoma de Manizales, Colombia), Fernanda Martínez Rodríguez (Universidad de Guadalajara, Mexico), Maria Jose Palancares Sosa (Instituto Politécnico Nacional, Mexico), Sonia H. Contreras-Ortiz (Universidad Tecnológica de Bolívar, Colombia), Simon Orozco-Arias (Universidad Autónoma de Manizales, Colombia), Mahmoud Hassaballah (South Valley University, Egypt), María de la Iglesia Vayá (Fundación para el Fomento de la Investigación Sanitario y Biomédica de la Comunidad Valenciana, Spain), Oscar Cardona-Morales (Universidad Autónoma de Manizales, Colombia), and Reinel Tabares-Soto (Universidad Autonóma de Manizales, Colombia)
Copyright: © 2022 |Pages: 46
DOI: 10.4018/978-1-6684-2304-2.ch007
OnDemand PDF Download:
No Current Special Offers


COVID-19 caused by the SARS-CoV-2 virus has affected healthcare and people's lifestyles worldwide since 2019. Among the available diagnostic tools, reverse transcription-polymerase chain reaction has proven highly accurate. However, the need for a specialized laboratory makes these tests expensive and time-consuming between sample collection and results. Currently, there are initial steps for the diagnosis of COVID-19 through chest x-ray images. Additionally, artificial intelligence techniques like deep learning (DL) help identify abnormalities. Inspired by the reported success of DL, this chapter presents an introduction to state-of-the-art DL-based approaches applied to the detection of COVID-19 in chest x-ray images, which currently allows assessing disease severity. The results presented are obtained using well-known models and some novel networks designed for this task. In addition, the models were evaluated using the most used public datasets, applying preprocessing techniques to improve detection results. Finally, this chapter shows some possible future research directions.
Chapter Preview


In Wuhan 2019, the emerged the coronavirus disease (COVID-19) (Roosa et al., 2020)⁠ caused by the virus severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) (Stoecklin et al., 2020), which is the third highest pathogenic coronavirus detected in humans preceded by the severe acute respiratory syndrome coronavirus (SARS-CoV) and the middle east respiratory syndrome coronavirus (MERS-CoV). On March 11, 2020, the World Health Organization (WHO) declared the COVID-19 outbreak a pandemic (Zhang et al., 2020)⁠. An infected person spreads the SARS-CoV-2 virus when coughs or sneezes primarily through droplets of saliva or discharge from the nose (World Health Organization, 2020). Infection can occur within 14 days of exposure but in most cases in no more than 4 to 5 days (Guan et al., 2020)⁠. The economical and health consequences of the rapid spread of the virus worldwide forced the adoption of extraordinary social confinement measures to halt its dissemination and to prevent the collapse of the health systems (Martínez Chamorro et al., 2021). It is estimated that between 30–40% of the COVID-19 patients are asymptomatic (Oran & Topol, 2020)⁠, others can experience broad symptoms severities. Most of the cases only presented mild or moderate symptoms (Yang et al., 2020)⁠, the most common are fever higher than 38°, cough, myalgia, smell and/or taste abnormalities, and headache (Martínez Chamorro et al., 2021). However, a study reports that 15.7% of patients admitted into the hospital developed severe illness (Guan et al., 2020)⁠, including severe pneumonia, pulmonary edema, acute respiratory distress syndrome, multiple organ failure, among others (Mo et al., 2020)⁠. The early identification and diagnosis are important to mitigate the COVID-19 outbreak (Guan et al., 2020), which as of July 16, 2021, the number of worldwide cases is 188,655,968, including 4,067,517 deaths (World Health Organization, 2021).

A report from the WHO solidarity consultum from February 2021 evaluates multiple drugs and shows that the mortality, initiation of ventilation, and hospitalization duration were not definitely reduced by any trial drug. Until now, no specific drug has been found against COVID-19 (WHO Solidarity Trial Consortium, 2021)⁠. Reverse transcription-polymerase chain reaction (RT-PCR) is among the most efficient and reliable methods for the detection of SARS-CoV-2, it is performed using a sample of nasopharyngeal or respiratory secretions. This method has a high specificity and sensitivity that can range from 60–70% to 95–97%, it can vary depending of the time elapsed since exposure to the virus (Martínez Chamorro et al., 2021)⁠. However, it takes hours to provide a result and needs specialized equipment and personal. The rapid antigen detection (RAD) tests are used to overcome these limitations since it is fast, easy to perform and interpret (Rathore & Ghosh, 2020)⁠.

Key Terms in this Chapter

COVID-19: A respiratory disease, caused by the SARS-CoV-2 virus, that spreads rapidly through saliva droplets or discharge from the nose. The most common symptoms are fever, dry cough, and fatigue.

Convolutional Neural Network (CNN): It is a type of deep learning model commonly used for image-related tasks. It uses the mathematical operation of convolution to extract features from images. In this chapter the models developed are based on CNN.

Artificial Intelligence: Artificial Intelligence refers to the set of algorithms or computational methods that aim to give computers the characteristics or abilities of human intelligence.

Semantic Segmentation: In the context of Deep Learning applications and image processing, semantic segmentation consists of assigning a label to each pixel in the image, which returns a segmentation mask for each input image. In this chapter, the labels for the semantic segmentation task are “Lung” and “Background”.

Classification: In the context of Deep Learning applications, a classification task consists of assigning a label to each sample of the input data based on the learned features and relationships. In this chapter, the classification task consists of assigning a Chest X-Ray image the label “Positive for COVID-19” or “Negative for COVID-19”.

Deep Learning: Deep Learning corresponds to a subset of Artificial Intelligence techniques that comprises models based on artificial neural networks.

Chest X-Ray: A Chest X-Ray is an image of the internal structures of the patient’s chest that is captured using radiation.

Complete Chapter List

Search this Book: