Computer assisted screening for lung cancer based on helical CT images

Describes a computer assisted automatic diagnosis system for lung cancer that detects tumor candidates at an early stage using helical computed tomography (CT) images. This automation of the process reduces the time and complexity and increases the confidence of diagnosis. The authors´ algorithm consists of an analysis part and a diagnosis part. In the analysis part, the authors´ method extracts the lung regions and the pulmonary blood vessels regions and analyzes the features of these regions using an image processing technique. In the diagnosis part, diagnostic rules are defined based on these features, and tumor candidates are detected using these rules. The authors apply their algorithm to the data of 450 mass screening patients. The results show that the algorithm detects lung cancer candidates successfully