Recent trends in Computer Aided diagnosis of lung nodules in thorax CT scans

摘要

The increased utilization of Computer Aided diagnosis (CAD) in clinical procedures has been very effective in discovering numerous abnormalities in human beings. CAD of lung nodules can be safely employed to validate the opinion of radiologists in discovering existence of nodules and assess the existence and severity of lung cancer. This paper provides a comprehensive review of the existing automated methods of identifying lung nodules found in thorax CT scans presented in the recent era. Information regarding characterization and size of nodules as well as nature of techniques (2D or 3D) is also incorporated in the paper. All the automated lung nodule detection schemes roughly entail few common procedures; i.e., pre-processing, lung or nodule segmentation, feature extraction and false positive reduction. The algorithms developed to execute all intermediate procedures are narrated. The paper presents comparison of the recent techniques in terms of the standard parameters employed by the researchers.