脊柱无创示踪装置在静态和术中动态环境下的临床精度研究

Objective To explore the accuracy performance of the newly designed non-invasive tracking system in static and dynamic environments. Methods The system consists of a non-invasive tracking device and an optical navigation device. In a static environment, the spinal model was scanned by intraoperative C-arm computed tomography (CT), and the marker points in the CT image space were selected for real-time automatic registration. The registration accuracy of the non-invasive tracking device and the registration accuracy of 8 body surface verification points were measured. The non-invasive tracking device was compared with the rigid tracking device connected to the vertebral body, and the position error of the vertebral bone surface point at 3-4 cm below the body surface and the position error of the vertebral internal point at 8-9 cm below the body surface were measured, as well as the angular error between the overall planned path of the guide needle and the actual path. In the previous study, we constructed a spinal vertebral motion model under respiratory motion to simulate the dynamic environment of the vertebral body during surgery. The dynamic environment simulates intraoperative breathing motion through robotic arm servo motion, and the dynamic environment accuracy measurement method was the same as that of the static environment. Results The non-invasive tracking device registration error was small, (0. 483 ± 0. 242) mm in static conditions and (0. 524±0. 229) mm in dynamic conditions. The registration error of the internal area of the non-invasive tracking device was lower than that of the external area of the non-invasive tracking device, [(0. 472 ± 0. 202) mm vs (0. 954 ± 0. 279) mm,P < 0. 001] in static conditions and [(0. 516±0. 188) mm vs (1. 029±0. 252) mm,P<0. 001] in dynamic conditions. The accuracy of the bone surface points and the bone internal points of the 8 guide needle approaches was related to the distance of the non-invasive tracking device. The position error at 3-4 cm below the body surface was smaller than that at 8-9 cm, [(0. 018±0. 024)mm vs (0. 061±0. 061) mm, P= 0. 002] in static conditions and [(0. 403±0. 275) mm vs (0. 938±0. 640) mm,P<0. 001] in dynamic conditions. The angle error of the needle approach is small, <0. 25° in static environment and <1. 50° in dynamic environment. Conclusions In a static environment, the noninvasive tracking device maintains high position accuracy and angular accuracy for both the surface and deep areas of the body. In a dynamic environment, the positioning accuracy and angular accuracy of the surface and deep areas of the body are reduced, but are still within an acceptable range.