System of computer-aided detection and positioning for lung nodules
肺结节计算机辅助检测与定位系统*

This paper established a system of computer-aided detection and positioning for lung nodules(SCADP). The SCADP system included function modules of segmentation of pulmonary parenchyma and candidate pulmonary nodules, selection and extraction of features, classification of lung nodules, image registration and fusion, 3D positioning of nodules, etc. It segmented the system candidate pulmonary nodules by GVF snake model, classified lung nodules based on rules and expert systems, realized the image registration by FFD, realized the multimodal image fusion by the fusion method based on wavelet transform with fusion rule of combining the local standard deviation and energy, and reconstructed 3D images by direct volume rendering based on improved Shear-Warp method. Experiments demonstrate that the SCADP providing a good means for computer-aided diagnosis of lung nodules and cancer.