A Systematic Survey of Automatic Detection of Lung Diseases from Chest X-Ray Images: COVID-19, Pneumonia, and Tuberculosis

Chest X-ray images (CXR) can convey a great deal about a patient’s condition; hence, the standard chest radiograph should be reconsidered. Interpretation of radiographs is challenging and requires skilled people to determine lung disease without false positives and negatives. A detailed investigation addressing lung diseases COVID-19, Pneumonia, and Tuberculosis is presented here with the goal of assisting investigators in constructing models that automatically identify lung diseases. This paper is presented in three folds. The first is an exploration of how research has progressed from classic feature engineering approaches to deep learning methods; the second is how these are used to identify the listed diseases using radiology images such as Chest X-rays (CXRs); and the third is the future path way of research to detect these diseases.