An AI Decision System to Predict Lung Nodules through Localization from Chest X-ray Images

The objective of this research is to design an Artificial Intelligence (AI) decision system that helps medical practitioners diagnose lung disease with more probability. Between the 1970s and 1990s, clinical image analysis was carried out by fusing low-level pixel processing (edge and line detectors, region expansion) with mathematical modeling (fitting lines, circles, and ellipses) to develop compound rule-based systems that dealt with particular tasks. In the 1990s the handwritten recognition task adapted Convolutional neural networks(CNN) and achieved the remarkable success that has laid the road for CNNs. The Chinese of Wuhan, in the Hubei Province, saw an epidemic of the coronavirus illness 2019 (COVID-19), which was unleashed by the SARS-CoV-2. The World Health organization declared the pandemic a public health emergency of worldwide significance on January 30, 2020. Over three years have passed since the worldwide struggle against the COVID-19 epidemic began. One of the quickest methods to diagnose patients is to use computer-aided diagnosis (CAD) and deep learning techniques on radiography and radiology images. With this, everyone’s focus has turned to lung disease detection such as tuberculosis and lung cancer through Chest X-ray images with state-of-the-art methods such as CNNs. However, these fail to explain the reason for making such a decision. This work focuses on this aspect of determining region of interest features with a process of localization while predicting lung nodules in two stages with the help of an explainable artificial method.