A Preoperative 3D Computer-Aided Diagnosis System for Lung Tumor

In recent years, the clinical treatment of computed tomography (CT) has become an important tool in medical imaging. Usually, we can integrate the computer-aided diagnosis (CAD) system to provide the physician the more reliable vision for preoperative diagnosing and analyzing. In this study, we aim to develop a novel region-based method for segmenting lung and lung tumors by the chest CT, and to obtain the information of lung such as its boundary, volume, localization of tumors. After the segmentation procedure, we also reconstruct the extracted lung and tumor regions to a three-dimensional (3D) volume module. In order to validate the performance of proposed CAD system, we have tested our method in a database from several lung patients. We also demonstrated the accuracy of the tumor segmentation method using a power statistical protocol. The experiment shows that our method results in higher accuracy in most cases.