在手术导航系统辅助的椎弓根螺钉内固定术中，术前与术中点云配准精度是影响导航定位效果的重要因素。由于术前与术中点云的获取方式不同，且术中暴露位置受限，两种非同源三维点云存在初始位姿差异大、重叠率低的问题，现有的配准算法在术前术中点云配准时容易失效且精度不高。为此，本文提出了一种椎弓根螺钉内固定术中非同源低重叠率点云的配准方法。首先，对术中点云进行降采样，对术前点云进行基于最远点采样的局部区域划分，选取其中的最优局部区域，与术中点云进行采样一致性初始配准；之后，采用迭代最近点算法进行进一步优化，实现点云的准确对齐。在本文所用数据集上，所提算法在术前术中点云配准实验中的平均旋转误差为0.406°，平移误差为0.474 mm，实现了初始位姿差异大、重叠率低的非同源术前术中点云的高精度配准，与现有算法相比，配准成功率从66.67%提高至100%。

Accurate placement of pedicle screw (PS) is crucial in spinal surgery. Developing new real-time intra-operative monitoring and navigation methods is an important direction of clinical application research. In this paper, we studied the spectrum along the fixation trajectory of PS in frequency domain to tackle the accuracy problem. Fresh porcine vertebrae, bovine vertebrae and ovine vertebrae were measured with the near-infrared spectrum (NIR) device to obtain the reflected spectrum from the vertebrae. Along the fixation trajectory of PS, average energy from different groups was calculated and used for identifying different tissues and compared to achieve the optimal recognition factor. Compared with the time domain approach, the frequency domain method could divide the spectra measured at different tissue points into different groups more stably and accurately, which could serve as a new method to assist the PS insertion. The results gained from this study are significant to the development of hi-tech medical instruments with independent intellectual property rights.

脊柱的治疗非常重要, 其中螺钉的准确植入是其内固定术的一个重要影响因素。 但由于椎骨外形差异较大, 很容易在螺钉植入过程中出现误植, 从而引起并发症。 因此, 需要寻找一种准确的椎弓根螺钉植入检测与监测方法。 提出了一种联合近红外参数和CT影像方法, 该方法可应用于椎弓根螺钉的植入过程中, 将计划植钉路径上椎骨组织的近红外参数和CT值进行研究, 提高术中植钉的准确率。 首先以猪椎骨为实验模型, 规划手术中的典型螺钉植入路径; 然后采用相关系统测量植入针道上的光谱, 计算光谱分析得到最佳近红外参数, 将其定义为特征因子; 进一步将规划路径上的CT值与特征因子进行关联, 建立三维空间CT值与近红外参数的转换模型。 结果表明CT值与特征因子具有线性关系, 为椎弓根螺钉植钉手术中的螺钉位置实时监测提供了参考依据, 该模型可以用于螺钉植入术中监测和误植预警。 论文提出的联合近红外参数和CT影像的椎弓根螺钉植入术中监测方法, 为临床手术中植钉准确率的提高奠定了基础。