目的:在医学影像领域,不同成像设备获取的信息不同,但互为补充,将这些信息融合在一起,为医生做出正确的诊断提供更多的依据,有助于提高医疗诊断和治疗水平.本文主要介绍医学图像配准技术的基本原理、方法、发展和临床应用.方法:大量参阅了该领域的有关文献,重点介绍了配准过程中的几何空间变换技术,分别综述了刚体变换和非刚体变换的主要方法,分析各种方法的优缺点,应用的适应范围,并对该领域的研究进行了展望.结果:在临床应用中刚体或近似刚体的配准只占很小的一部分,大多数情况下需要用非刚体变换来描述.目前,非刚体变换方法还不够成熟,各类方法都存在一定的局限性,在临床实际应用中,根据研究对象的差别建立合理的配准形变模型,选择合适的配准方法,优化计算方法,提高计算速度,进一步提高图像配准的准确性.结论:论文认为医学图像配准技术是医学图像处理研究领域中的热点方向之一,具有重要的临床应用价值.论文提出图像配准的精度直接影响后续的图像处理工作,如何提高配准精度仍是今后医学图像配准技术的研究方向.%Objective: In the field of medical imaging, the information provided by various image instruments is different, but complementary. Putting all information together provides more evidence to the doctor making right diagnosis, also helps improve medical diagnosis and treatment level. This paper mainly describes the basic principles, methods, development and clinical applications of image registration. Methods: Based on a lot of the literature in the field, the paper mainly introduces the technology of geometric space transformation registration, and summarizes the rigid transformation and non-rigid transformation's methods, analyzes the advantages and disadvantages of these methods, application adaptation.And makes a prospects about the research in this field. Results: In the clinical application, rigid bodies is limited,In most cases the non-rigid transformation is needed. Presently, Non-rigid transformation methods are not mature enough, various methods have certain limitations. In clinical practical applications, establishing reasonable registration deformation model according to the different object, selecting the appropriate methods optimizing calculation methods, improving the calculating speed will further improve the accuracy of image registration. Conclusions: The article holds the opinion that the medical image registration technology is one of the hot spots in the field of medical image processing research, possessing important clinical value. This paper puts forward the accuracy of image registration directly affects the subsequent work of image processing. How to improve registration accuracy is still a research direction in future medical image registration technology.
展开▼
