Digital Medical Research Centre, Fudan University, Shanghai, People's Republic of China.
Int J Med Robot. 2010 Jun;6(2):142-9. doi: 10.1002/rcs.299.
Point-pair registration is widely used to register the patient and the image space in image-guided neurosurgery. The registration accuracy at the target point is influenced by the distribution of the fiducial points, and it is not always easy to achieve a good distribution in clinical practice.
We propose one original configuration and three variants of the distribution of the fiducial points, in which each fiducial point has a definite position on the head surface. These configurations are conceived by considering both the theoretical value of the target registration error (TRE) and practical constraints on the positioning of the fiducial points. These configurations can guide the surgeon to place fiducial points directly onto the patient's head.
Experiments with real clinical data showed that when using the original configuration, the TRE in the whole skull was < 2 mm and the TRE in the whole brain was < 1.5 mm if the fiducial registration error was 2.5 mm. When using variants of the original configuration, small TRE can also be achieved in the corresponding surgical field.
This study provides an easy-to-use approach to achieving good distribution of fiducial points in image-guided neurosurgery.
点对配准广泛用于在图像引导神经外科中对患者和图像空间进行配准。目标点处的配准精度受基准点分布的影响，在临床实践中并不总是容易实现良好的分布。
我们提出了一种原始配置和三种基准点分布的变体，其中每个基准点在头部表面上都有确定的位置。这些配置是通过考虑目标注册误差（TRE）的理论值和基准点定位的实际约束来构思的。这些配置可以指导外科医生将基准点直接放置在患者的头部上。
使用真实临床数据进行的实验表明，使用原始配置时，如果基准点注册误差为 2.5 毫米，则整个颅骨中的 TRE<2 毫米，整个大脑中的 TRE<1.5 毫米。使用原始配置的变体时，在相应的手术区域也可以实现较小的 TRE。
本研究为在图像引导神经外科中实现良好的基准点分布提供了一种易于使用的方法。
