Tu Puxun, Wang Huixiang, Joskowicz Leo, Chen Xiaojun
Institute of Biomedical Manufacturing and Life Quality Engineering, School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai, China.
Department of Orthopedics, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China.
Int J Comput Assist Radiol Surg. 2023 Sep;18(9):1715-1724. doi: 10.1007/s11548-023-02884-4. Epub 2023 Apr 8.
The treatment of pelvic and acetabular fractures remains technically demanding, and traditional surgical navigation systems suffer from the hand-eye mis-coordination. This paper describes a multi-view interactive virtual-physical registration method to enhance the surgeon's depth perception and a mixed reality (MR)-based surgical navigation system for pelvic and acetabular fracture fixation.
First, the pelvic structure is reconstructed by segmentation in a preoperative CT scan, and an insertion path for the percutaneous LC-II screw is computed. A custom hand-held registration cube is used for virtual-physical registration. Three strategies are proposed to improve the surgeon's depth perception: vertices alignment, tremble compensation and multi-view averaging. During navigation, distance and angular deviation visual cues are updated to help the surgeon with the guide wire insertion. The methods have been integrated into an MR module in a surgical navigation system.
Phantom experiments were conducted. Ablation experimental results demonstrated the effectiveness of each strategy in the virtual-physical registration method. The proposed method achieved the best accuracy in comparison with related works. For percutaneous guide wire placement, our system achieved a mean bony entry point error of 2.76 ± 1.31 mm, a mean bony exit point error of 4.13 ± 1.74 mm, and a mean angular deviation of 3.04 ± 1.22°.
The proposed method can improve the virtual-physical fusion accuracy. The developed MR-based surgical navigation system has clinical application potential. Cadaver and clinical experiments will be conducted in future.
骨盆和髋臼骨折的治疗在技术上仍具有挑战性，传统手术导航系统存在手眼不协调问题。本文描述了一种多视图交互式虚拟-物理配准方法，以增强外科医生的深度感知，并介绍了一种基于混合现实（MR）的骨盆和髋臼骨折固定手术导航系统。
首先，在术前CT扫描中通过分割重建骨盆结构，并计算经皮LC-II螺钉的插入路径。使用定制的手持式配准立方体进行虚拟-物理配准。提出了三种策略来提高外科医生的深度感知：顶点对齐、抖动补偿和多视图平均。在导航过程中，更新距离和角度偏差视觉提示，以帮助外科医生插入导丝。这些方法已集成到手术导航系统的MR模块中。
进行了体模实验。消融实验结果证明了虚拟-物理配准方法中各策略的有效性。与相关工作相比，该方法具有最佳的准确性。对于经皮导丝置入，我们的系统实现了平均骨入口点误差为2.76±1.31mm，平均骨出口点误差为4.13±1.74mm，平均角度偏差为3.04±1.22°。
所提出的方法可以提高虚拟-物理融合精度。所开发的基于MR的手术导航系统具有临床应用潜力。未来将进行尸体和临床实验。
