Computer-aided diagnosis of lung cancers based on quantitative analysis of time-interval changes on thoracic CT images

点击翻译按钮获取中文摘要

Lung cancer is the most common cause of cancer death worldwide. CT is the modality of choice for lung imaging because this modality is the most sensitive way for imaging lung in vivo currently. Some evidences suggested that CT with a low-dose imaging protocol may improve pulmonary nodule sensitivity relative to radiography. While this study does show the efficacy of CT based screening, physicians often face the problem of deciding an appropriate management strategies for maximizing patient survival and preserving lung function. This study focuses on a framework within which CT value histograms is used to represent internal structures of pulmonary nodules, aiming at classifying patients between those at high risk of disease progression (high-risk group) and those at low risk of disease progression (low-risk group).