摘要
目的:尝试一种基于体表定位的二维图像配准方法,实现"正电子发射断层成像"(PT)和"计算机断层扫描"(CT)异机图像的精确融合。方法:输入PT/CT原始数据后采用数字化格式转换,设计"9点3面"立体定位法进行配准,在实时工作站Mimics按照信息交互自动融合模式,通过讯号叠加技术完成图像融合结果:以肺癌患者的躯干冠状面为实例试验[PT+CT]二维图像的异机融合,生成了同时呈现躯干结构和代谢状况的互补影像,快捷而清晰。结论:图像融合具有明确放射诊断以及量化治疗方案的临床意义;在[PT+CT]联机设备尚未普及的情况下,异机融合是同机融合功用的技术补充。
Objective：To explore for a localization registration scheme of 2-dimension （2D） image based on somatotopic localization and to achieve accurate cross-modality image fusion of positron emission tomography （PET） and computed tomography. Methods：The raw data of PET/CT were converted into digital formats. Using a designed ＂9-point and 3-plane＂ somatotopic localization scheme, PET-CT image fusion was completed with superposition of signals by interactive auto-fusing mode at a real-time workstation （Mimics）. Results：The hybrid PET/CT image fusion was attempted for the coronal trunk scanning in patients with lung cancer,which resulted in quick and clear display of anatomical structure and metabolic status in one single hybrid image. Conelusion： Image fusion can be clinically relevant for a confirmed radiologic diagnosis and quantitative treatment strategy. Cross-modality image fusion may be used as a technical complement until single-modality PET/CT imaging becomes widely available.
出处
《广州医学院学报》
2011年第3期1-4,35,共5页
Academic Journal of Guangzhou Medical College
基金
广州医学院2009年度留学回国
博士启动基金项目(20091206)
广州人事局2010年度广州留学人员科技创业专项基金项目(20100804)
关键词
电成像
核磁共振
计算机断层扫描
定位配准
异机图像融合
positron emission tomography （PET）
magnetic resonance imaging （MRI）
X-ray computed tomography （CT）
localization registration
cross modality image fusion
