A novel machine learning‐based analytical framework for automatic detection of COVID‐19 using chest X‐ray images. Issue 3 (11th June 2021)

• Record Type: Journal Article
• Authors: Johri, Shikhar; Goyal, Mehendi; Jain, Sahil; Baranwal, Manoj; Kumar, Vinay; Upadhyay, Rahul
• Is Part Of: International journal of imaging systems and technology. Volume 31:Issue 3(2021)
• Page Start: 1105
• Page End: 1119
• Publication Date: 2021-06-11
• Subjects: chest X‐ray images -- coronavirus -- machine learning methods -- pneumonia; Imaging systems -- Periodicals; Image processing -- Periodicals; 621.367
• DOI: 10.1002/ima.22613
• Languages: English
• ISSNs: 0899-9457

Abstract

Abstract: Considering the prevailing scenario of COVID‐19 pandemic, early detection of the disease is an important and crucial step in disease management. Early detection and correct treatment may limit disease progression to severe levels and prevent deaths. In addition, early isolation of infected patients will lead to control transmission rate and will possibly reduce the stress on the present healthcare system. Currently, the most common and reliable testing method available for COVID‐19 diagnosis is real‐time reverse transcription‐polymerase chain reaction (rRT‐PCR) test. However, the chest radiological (X‐ray) imaging can be used as an alternate method to rRT‐PCR test, and early COVID‐19 symptoms can be investigated by critical examination of patient's chest scans. In the present work, a novel machine learning (ML)‐based analytical framework is developed for automatic detection of COVID‐19 using chest X‐ray (CXR) images of plausible patients. The framework is designed, trained, and validated to identify four classes of CXR images namely, healthy, bacterial pneumonia, viral pneumonia, and COVID‐19. The experimental results pose the proposed framework as a potential candidate for COVID‐19 disease diagnosis using CXR images, with training, validation, and testing accuracy of 92.4%, 88.24%, and 87.13%, respectively, in four‐class classification. The comparative analysis demonstrates the better capabilities of the proposed framework COVID‐19 detection along with other types of pneumonia.