摘要
介绍了用于开发肺部CT图像计算机辅助诊断(computer aided diagnosis,CAD)系统的LIDC/IDRI(Lung Image Database Consortium/Image Database Resource Initiative)影像数据库的发展历史、文件格式、组织构架及基本应用,分析了基于该影像数据库的肺结节分割与诊断的研究进展。指出了LIDC/IDRI影像数据库与CAD系统之间的关系,并指明了基于LIDC/IDRI影像数据库的新的人工智能方案,如深层强化学习和迁移学习等,是肺结节CAD系统未来的发展方向。
Lung hnage Database Consortium (LIDC) / Image Database Resource Initiative (IDRI) image database had its history, file format, organizational structure and application introduced when used for developing a lung CT image computer- aided diagnosis (CAD) system. The research progress of pulmonary nodule segmentation and diagnosis based on the image database was analyzed, and the relationship between LIDC/IDRI image database and CAD system was indicated. It's pointed out that new AI schedules based on LIDC/IDRI image database, such as deep reinforcement learning and transfer learning, would contribute to the development of puhnonary nodule CAD system.
作者
林岚
吴玉超
宋爽
吴水才
LIN Lan;WU Yu-chao;SONG Shuang;WU Shui-cai(College of Life Science and Bioengineering,Beijing University of Technology,Beijing 100124,China)
出处
《医疗卫生装备》
CAS
2018年第10期95-99,共5页
Chinese Medical Equipment Journal
基金
北京市教委科技计划一般项目(KM201810005033)
国家科技支撑计划课题(2015BAI02B03)