Abstract:

Abstract Lung cancer is one of the major killers, hence early cancer identification is crucial to improve survival chances. The most popular option for early screening and identifying lung illnesses is a computed tomography (CT) scan. However, even for seasoned radiologists, manual lung disease detection and labeling requires a lot of time and effort since a sophisticated CT scanner generates a lot of CT images. An automated Computer-Aided Diagnosis (CAD) of pulmonary CT scan to aid the Radiologists is a workable answer to this. The primary topic of this paper is the development of deep learning algorithm-based methods for the early prediction of lung malignancies, which are carried out in three stages: (a) lung segmentation; (b) classification of interstitial lung disease (ILD); and (c) classification of lung cancer. The strength of the overall CAD system for diagnosing lung disorders depends on precise lung segmentation throughout the multi-stage process of automatic evaluation of the lung CT image.

Keywords: