肺结节计算机辅助检测与定位系统  

System of computer-aided detection and positioning for lung nodules

作　　者：李彬[1] 欧陕兴[1] 田联房[2] 齐燕[1] 刘思伟[2] 张婧[2] 

机构地区：[1]广州军区广州总医院放射科,广州510010 [2]华南理工大学自动化科学与工程学院,广州510640

出　　处：《计算机应用研究》2010年第6期2184-2188,共5页Application Research of Computers

基　　金：中国博士后基金资助项目(20090450866);广东省自然科学基金资助项目(8451064101000631);国家教育部博士点基金资助项目(200805610018);广州市番禺区科技攻关项目(2009-Z-108-1)

摘　　要：为了实现肺结节的智能识别,开发了肺结节计算机辅助检测与定位系统(SCADP)。该系统包括肺实质分割、候选肺结节分割、肺结节特征选择与提取、肺结节分类、图像配准与融合、结节三维重建与定位和勾画病灶等功能模块。其中,采用活动轮廓模型的分割方法实现候选肺结节分割,采用基于规则与专家系统的决策方法实现肺结节分类;采用自由变形法实现图像配准。基于小波变换的融合方法,以区域标准差与区域能量相结合的融合规则实现多模图像的融合;基于改进的Shear-Warp算法快速实现体绘制。实验证明,该肺结节计算机辅助检测与定位系统满足肺结节计算机辅助诊断要求。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.

关 键 词：肺结节 计算机辅助检测 分类 可视化 系统功能模块 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]