Pulmonary organs analysis for differential diagnosis based on thoracic thin-section CT images

摘要

The pulmonary organs have a very complicated structure which consists of the bronchus, the pulmonary artery, and the pulmonary vein. So it is difficult for a medical doctor to understand the spatial relationships among a tumor and the pulmonary organs. Here, the authors present a 3D image analysis method of the pulmonary organs based on thin-section CT images. First, they extract the regions of interest using Gaussian curvature and mean curvature of lung field, then they classify the bronchus and the blood vessels, and they conclude with the identification of the pulmonary artery and the pulmonary vein. The authors use this method to support a differential diagnosis of lung cancer by making quantitative analysis of abnormal nodules.