作者: Marco Feuerstein , Ben Glocker , Takayuki Kitasaka , Yoshihiko Nakamura , Shingo Iwano
DOI: 10.1016/J.MEDIA.2011.05.005
关键词:
摘要: One important aspect of lung cancer staging is the assessment mediastinal lymph nodes in 3-D chest computed tomography (CT) images. In current clinical routine this done manually by analyzing CT image slice to find nodes, evaluate them quantitatively, and assign labels for describing pathologic extent metastases. paper we present a method automate process node detection labeling creation average novel atlas containing probability maps mediastinal, aortic, N1 nodes. Utilizing fast deformable registration approach match with images new patients, our can maintain an acceptable runtime. comparison previously published methods it also shows good sensitivity positive predictive value.